Improve selftest c_ctrl_unctrl
Checks
| Context |
Check |
Description |
| linaro-tcwg-bot/tcwg_gdb_build--master-aarch64 |
success
|
Build passed
|
| linaro-tcwg-bot/tcwg_gdb_build--master-arm |
success
|
Build passed
|
| linaro-tcwg-bot/tcwg_gdb_check--master-arm |
success
|
Test passed
|
| linaro-tcwg-bot/tcwg_gdb_check--master-aarch64 |
success
|
Test passed
|
Commit Message
Improve selftest c_ctrl_unctrl by exercising all valid inputs of c_iscntrl.
Tested on x86_64-linux.
Suggested-By: Tom Tromey <tom@tromey.com> [1]
[1] https://sourceware.org/pipermail/gdb-patches/2026-March/226231.html
---
gdb/utils.c | 24 +++++++++++++++++++++---
1 file changed, 21 insertions(+), 3 deletions(-)
base-commit: edf0ae3b9f1ca23daaa55628683edc9a3360286f
Comments
On 3/26/26 3:04 PM, Tom de Vries wrote:
> Improve selftest c_ctrl_unctrl by exercising all valid inputs of c_iscntrl.
>
Ping.
Thanks,
- Tom
> Tested on x86_64-linux.
>
> Suggested-By: Tom Tromey <tom@tromey.com> [1]
>
> [1] https://sourceware.org/pipermail/gdb-patches/2026-March/226231.html
> ---
> gdb/utils.c | 24 +++++++++++++++++++++---
> 1 file changed, 21 insertions(+), 3 deletions(-)
>
> diff --git a/gdb/utils.c b/gdb/utils.c
> index b073bb9fdf3..c8903896081 100644
> --- a/gdb/utils.c
> +++ b/gdb/utils.c
> @@ -3793,12 +3793,30 @@ test_c_ctrl_unctrl ()
> SELF_CHECK (c_ctrl ('?') == 0x7f);
> SELF_CHECK (c_unctrl (0x7f) == '?');
>
> - /* Consistency check. */
> - for (unsigned int i = 0; i < 0x100; i++)
> + /* Consistency check.
> +
> + The ctype.h function iscntrl has an int parameter, and defined behavior
> + for inputs [0, 255] and EOF. Consequently, iscntrl has undefined
> + behavior when called with a negative signed char argument (with the
> + possible exception of EOF, which is a negative int constant). That
> + extends to chars on platforms where char == signed char. Consequently,
> + to write portable code, iscntrl char and signed char arguments need to be
> + cast to unsigned char.
> +
> + The c-ctype.h variant c_iscntrl sidesteps this problem by having defined
> + behavior for inputs [-128, 255] and not bothering about EOF.
> +
> + The functions c_ctrl/c_unctrl sidestep this issue by using parameter
> + type unsigned char instead of int.
> +
> + So, testing input values [0..255] tests all relevant behavior of
> + c_ctrl/c_unctrl. However, we designed these functions to be compatible
> + with c_iscntrl, so we test for all valid inputs of that function. */
> + for (int i = -128; i < 256; i++)
> {
> unsigned char ch = i;
> unsigned char unctrl_ch = c_unctrl (ch);
> - if (!c_iscntrl (ch))
> + if (!c_iscntrl (i))
> {
> SELF_CHECK (unctrl_ch == ch);
> continue;
>
> base-commit: edf0ae3b9f1ca23daaa55628683edc9a3360286f
@@ -3793,12 +3793,30 @@ test_c_ctrl_unctrl ()
SELF_CHECK (c_ctrl ('?') == 0x7f);
SELF_CHECK (c_unctrl (0x7f) == '?');
- /* Consistency check. */
- for (unsigned int i = 0; i < 0x100; i++)
+ /* Consistency check.
+
+ The ctype.h function iscntrl has an int parameter, and defined behavior
+ for inputs [0, 255] and EOF. Consequently, iscntrl has undefined
+ behavior when called with a negative signed char argument (with the
+ possible exception of EOF, which is a negative int constant). That
+ extends to chars on platforms where char == signed char. Consequently,
+ to write portable code, iscntrl char and signed char arguments need to be
+ cast to unsigned char.
+
+ The c-ctype.h variant c_iscntrl sidesteps this problem by having defined
+ behavior for inputs [-128, 255] and not bothering about EOF.
+
+ The functions c_ctrl/c_unctrl sidestep this issue by using parameter
+ type unsigned char instead of int.
+
+ So, testing input values [0..255] tests all relevant behavior of
+ c_ctrl/c_unctrl. However, we designed these functions to be compatible
+ with c_iscntrl, so we test for all valid inputs of that function. */
+ for (int i = -128; i < 256; i++)
{
unsigned char ch = i;
unsigned char unctrl_ch = c_unctrl (ch);
- if (!c_iscntrl (ch))
+ if (!c_iscntrl (i))
{
SELF_CHECK (unctrl_ch == ch);
continue;