| Message ID | 20170511162127.15692-1-luc.vanoostenryck@gmail.com (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Headers | show |
On Thu, May 11, 2017 at 9:21 AM, Luc Van Oostenryck <luc.vanoostenryck@gmail.com> wrote: > Conversion to bool is special in C since this conversion > is essentially the result of the comparison with zero. > As such, some operations which are normally unsafe to > do with restricted types, like casting to an unrestricted > type, are in fact safe to do when converting to bool > and issuing a warning in those case is useless, confusing > and causes people to add useless casts in the code in > order to shut up the warning. > > Fix this by catching such 'bool <- restricted type' conversion > and avoid such warnings. The change seems fine. However, have you run the test-suite with this change? It seems cause some test do not pass. I assume it is cause by different error message it will output. Chris -- To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, May 11, 2017 at 9:41 PM, Christopher Li <sparse@chrisli.org> wrote: > On Thu, May 11, 2017 at 9:21 AM, Luc Van Oostenryck > <luc.vanoostenryck@gmail.com> wrote: >> Conversion to bool is special in C since this conversion >> is essentially the result of the comparison with zero. >> As such, some operations which are normally unsafe to >> do with restricted types, like casting to an unrestricted >> type, are in fact safe to do when converting to bool >> and issuing a warning in those case is useless, confusing >> and causes people to add useless casts in the code in >> order to shut up the warning. >> >> Fix this by catching such 'bool <- restricted type' conversion >> and avoid such warnings. > > The change seems fine. However, have you run the test-suite > with this change? It seems cause some test do not pass. > I assume it is cause by different error message it will output. Oh my, sorry. I *have* run the test-suite but then I've forgotten to commit the change needed for two test cases. Thanks for noticing this. -- Luc -- To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/evaluate.c b/evaluate.c index 976857915..3dc26fc09 100644 --- a/evaluate.c +++ b/evaluate.c @@ -1345,6 +1345,12 @@ static int check_assignment_types(struct symbol *target, struct expression **rp, return 1; } else if (!(sclass & TYPE_RESTRICT)) goto Cast; + if (t == &bool_ctype) { + if (is_fouled_type(s)) + warning((*rp)->pos, "%s degrades to integer", + show_typename(s->ctype.base_type)); + goto Cast; + } *typediff = "different base types"; return 0; } diff --git a/validation/bool-cast-restricted.c b/validation/bool-cast-restricted.c new file mode 100644 index 000000000..f6776b050 --- /dev/null +++ b/validation/bool-cast-restricted.c @@ -0,0 +1,25 @@ +typedef unsigned int __attribute__((bitwise)) large_t; +#define LBIT ((__attribute__((force)) large_t) 1) + +_Bool lfoo(large_t x) { return x; } +_Bool lbar(large_t x) { return ~x; } +_Bool lbaz(large_t x) { return !x; } +_Bool lqux(large_t x) { return x & LBIT; } + + +typedef unsigned short __attribute__((bitwise)) small_t; +#define SBIT ((__attribute__((force)) small_t) 1) + +_Bool sfoo(small_t x) { return x; } +_Bool sbar(small_t x) { return ~x; } +_Bool sbaz(small_t x) { return !x; } +_Bool squx(small_t x) { return x & SBIT; } + +/* + * check-name: bool-cast-restricted.c + * check-command: sparse -Wno-decl $file + * + * check-error-start +bool-cast-restricted.c:14:32: warning: restricted small_t degrades to integer + * check-error-end + */
Conversion to bool is special in C since this conversion is essentially the result of the comparison with zero. As such, some operations which are normally unsafe to do with restricted types, like casting to an unrestricted type, are in fact safe to do when converting to bool and issuing a warning in those case is useless, confusing and causes people to add useless casts in the code in order to shut up the warning. Fix this by catching such 'bool <- restricted type' conversion and avoid such warnings. CC: Al Viro <viro@zeniv.linux.org.uk> Originally-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> --- evaluate.c | 6 ++++++ validation/bool-cast-restricted.c | 25 +++++++++++++++++++++++++ 2 files changed, 31 insertions(+) create mode 100644 validation/bool-cast-restricted.c