As the name shows, it checks if strings are equal in case insensitive
manner.
For example, drivers/infiniband/ulp/rtrs/rtrs-clt-sysfs.c uses
strncasecmp to check that the input via sysfs is "mi". But it would
work even-if the input is "min-wrongcommand".
I found some more cases using strncasecmp to check the entire string
such as rtrs-clt-sysfs.c does. drivers/pnp/interface.c checks
"disable" command with strncasecmp but it would also work if the
command is "disable-wrong".
Signed-off-by: Gioh Kim <[email protected]>
---
include/linux/string.h | 1 +
lib/string.c | 23 +++++++++++++++++++++++
2 files changed, 24 insertions(+)
diff --git a/include/linux/string.h b/include/linux/string.h
index 4fcfb56abcf5..36d00ff8013e 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -184,6 +184,7 @@ extern char **argv_split(gfp_t gfp, const char *str, int *argcp);
extern void argv_free(char **argv);
extern bool sysfs_streq(const char *s1, const char *s2);
+extern bool sysfs_streqcase(const char *s1, const char *s2);
extern int kstrtobool(const char *s, bool *res);
static inline int strtobool(const char *s, bool *res)
{
diff --git a/lib/string.c b/lib/string.c
index 7548eb715ddb..5e6bc0d3d5c6 100644
--- a/lib/string.c
+++ b/lib/string.c
@@ -714,6 +714,29 @@ bool sysfs_streq(const char *s1, const char *s2)
}
EXPORT_SYMBOL(sysfs_streq);
+/**
+ * sysfs_streqcase - same to sysfs_streq and case insensitive
+ * @s1: one string
+ * @s2: another string
+ *
+ */
+bool sysfs_streqcase(const char *s1, const char *s2)
+{
+ while (*s1 && tolower(*s1) == tolower(*s2)) {
+ s1++;
+ s2++;
+ }
+
+ if (*s1 == *s2)
+ return true;
+ if (!*s1 && *s2 == '\n' && !s2[1])
+ return true;
+ if (*s1 == '\n' && !s1[1] && !*s2)
+ return true;
+ return false;
+}
+EXPORT_SYMBOL(sysfs_streqcase);
+
/**
* match_string - matches given string in an array
* @array: array of strings
--
2.25.1
On Tue, Apr 6, 2021 at 11:15 PM Gioh Kim <[email protected]> wrote:
>
> As the name shows, it checks if strings are equal in case insensitive
> manner.
>
> For example, drivers/infiniband/ulp/rtrs/rtrs-clt-sysfs.c uses
> strncasecmp to check that the input via sysfs is "mi". But it would
> work even-if the input is "min-wrongcommand".
>
> I found some more cases using strncasecmp to check the entire string
> such as rtrs-clt-sysfs.c does. drivers/pnp/interface.c checks
> "disable" command with strncasecmp but it would also work if the
> command is "disable-wrong".
Reviewed-by: Nick Desaulniers <[email protected]>
I do wonder if these (sysfs_streqcase and sysfs_streq) could or should
be conditionally available on CONFIG_SYSFS=y; don't pay for what you
don't use (without needing CONFIG_LD_DEAD_CODE_DATA_ELIMINATION=y)?
Also, it might be nice to share the second half of the function with
sysfs_streq via a new static function. Though it will just get
inlined in both for CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y, it might
help the compiler if CONFIG_CC_OPTIMIZE_FOR_SIZE=y was instead chosen
if the compiler cannot outline/deduplicate the shared code. At the
least, there's less duplication between two very similar functions; if
one changes then authors may need to be careful to update both.
Are either of those concerns worth a v3? ¯\_(ツ)_/¯
>
> Signed-off-by: Gioh Kim <[email protected]>
> ---
> include/linux/string.h | 1 +
> lib/string.c | 23 +++++++++++++++++++++++
> 2 files changed, 24 insertions(+)
>
> diff --git a/include/linux/string.h b/include/linux/string.h
> index 4fcfb56abcf5..36d00ff8013e 100644
> --- a/include/linux/string.h
> +++ b/include/linux/string.h
> @@ -184,6 +184,7 @@ extern char **argv_split(gfp_t gfp, const char *str, int *argcp);
> extern void argv_free(char **argv);
>
> extern bool sysfs_streq(const char *s1, const char *s2);
> +extern bool sysfs_streqcase(const char *s1, const char *s2);
> extern int kstrtobool(const char *s, bool *res);
> static inline int strtobool(const char *s, bool *res)
> {
> diff --git a/lib/string.c b/lib/string.c
> index 7548eb715ddb..5e6bc0d3d5c6 100644
> --- a/lib/string.c
> +++ b/lib/string.c
> @@ -714,6 +714,29 @@ bool sysfs_streq(const char *s1, const char *s2)
> }
> EXPORT_SYMBOL(sysfs_streq);
>
> +/**
> + * sysfs_streqcase - same to sysfs_streq and case insensitive
> + * @s1: one string
> + * @s2: another string
> + *
> + */
> +bool sysfs_streqcase(const char *s1, const char *s2)
> +{
> + while (*s1 && tolower(*s1) == tolower(*s2)) {
> + s1++;
> + s2++;
> + }
> +
> + if (*s1 == *s2)
> + return true;
> + if (!*s1 && *s2 == '\n' && !s2[1])
> + return true;
> + if (*s1 == '\n' && !s1[1] && !*s2)
> + return true;
> + return false;
> +}
> +EXPORT_SYMBOL(sysfs_streqcase);
> +
> /**
> * match_string - matches given string in an array
> * @array: array of strings
> --
> 2.25.1
>
--
Thanks,
~Nick Desaulniers
On Wed, Apr 7, 2021 at 10:07 PM Nick Desaulniers
<[email protected]> wrote:
>
> On Tue, Apr 6, 2021 at 11:15 PM Gioh Kim <[email protected]> wrote:
> >
> > As the name shows, it checks if strings are equal in case insensitive
> > manner.
> >
> > For example, drivers/infiniband/ulp/rtrs/rtrs-clt-sysfs.c uses
> > strncasecmp to check that the input via sysfs is "mi". But it would
> > work even-if the input is "min-wrongcommand".
> >
> > I found some more cases using strncasecmp to check the entire string
> > such as rtrs-clt-sysfs.c does. drivers/pnp/interface.c checks
> > "disable" command with strncasecmp but it would also work if the
> > command is "disable-wrong".
>
> Reviewed-by: Nick Desaulniers <[email protected]>
>
> I do wonder if these (sysfs_streqcase and sysfs_streq) could or should
> be conditionally available on CONFIG_SYSFS=y; don't pay for what you
> don't use (without needing CONFIG_LD_DEAD_CODE_DATA_ELIMINATION=y)?
Good idea.
Thank you.
>
> Also, it might be nice to share the second half of the function with
> sysfs_streq via a new static function. Though it will just get
> inlined in both for CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y, it might
> help the compiler if CONFIG_CC_OPTIMIZE_FOR_SIZE=y was instead chosen
> if the compiler cannot outline/deduplicate the shared code. At the
> least, there's less duplication between two very similar functions; if
> one changes then authors may need to be careful to update both.
Yes, they are exactly the same.
I will make an inline function for the common code.
>
> Are either of those concerns worth a v3? ¯\_(ツ)_/¯
Sure, I will not forget to add 'V2'.
Thank you for kind review.