On Tue, Oct 20, 2015 at 3:55 PM, Ilya Kaliman <ilya.kali...@gmail.com> wrote:
...
> --- bin/ed/main.c 9 Oct 2015 21:24:05 -0000 1.53
> +++ bin/ed/main.c 20 Oct 2015 22:49:53 -0000
> @@ -174,7 +174,7 @@ top:
> signal(SIGHUP, signal_hup);
> signal(SIGQUIT, SIG_IGN);
> signal(SIGINT, signal_int);
> - if (status = sigsetjmp(env, 1)) {
> + if ((status = sigsetjmp(env, 1))) {
This technically undefined behavior, as you are not allowed to capture
the return value of setjmp(), _setjmp(), or sigsetjmp() in a variable
To quote the C standard:
----
An application shall ensure that an invocation of setjmp( ) appears in
one of the following
contexts only:
· The entire controlling expression of a selection or iteration statement
· One operand of a relational or equality operator with the other
operand an integral
constant expression, with the resulting expression being the
entire controlling expression
of a selection or iteration statement
· The operand of a unary '!' operator with the resulting
expression being the entire
controlling expression of a selection or iteration
· The entire expression of an expression statement (possibly cast to void)
If the invocation appears in any other context, the behavior is undefined.
----
POSIX extends that to the other two functions.
Fortunately, ed only call sigsetjmpt() with the value -1, and status
is initialized to zero, so the diff below should fix it without
changing the behavior.
oks?
Philip Guenther
--- bin/ed/main.c 21 Oct 2015 16:06:57 -0000 1.54
+++ bin/ed/main.c 23 Oct 2015 04:55:31 -0000
@@ -174,7 +174,8 @@ top:
signal(SIGHUP, signal_hup);
signal(SIGQUIT, SIG_IGN);
signal(SIGINT, signal_int);
- if ((status = sigsetjmp(env, 1))) {
+ if (sigsetjmp(env, 1)) {
+ status = -1;
fputs("\n?\n", stderr);
seterrmsg("interrupt");
} else {
