[U-Boot] [PATCH 4/4] dm: simplify the loop in lists_driver_lookup_name()

[Igor Grinberg]

Hi Masahiro, Simon,

On 09/28/14 18:22, Simon Glass wrote:
> Hi Masahiro,
> 
> On 28 September 2014 07:52, Masahiro Yamada <yamada.m at jp.panasonic.com> wrote:
>>         if (strncmp(name, entry->name, len))
>>                 continue;
>>
>>         /* Full match */
>>         if (len == strlen(entry->name))
>>                 return entry;
>>
>> is equivalent to:
>>
>>         if (!strcmp(name, entry->name))
>>                 return entry;
>>
>> The latter is simpler.
>>
>> Signed-off-by: Masahiro Yamada <yamada.m at jp.panasonic.com>
>> ---
>>
>>  drivers/core/lists.c | 9 +--------
>>  1 file changed, 1 insertion(+), 8 deletions(-)
>>
>> diff --git a/drivers/core/lists.c b/drivers/core/lists.c
>> index 699f94b..3a1ea85 100644
>> --- a/drivers/core/lists.c
>> +++ b/drivers/core/lists.c
>> @@ -24,19 +24,12 @@ struct driver *lists_driver_lookup_name(const char *name)
>>                 ll_entry_start(struct driver, driver);
>>         const int n_ents = ll_entry_count(struct driver, driver);
>>         struct driver *entry;
>> -       int len;
>>
>>         if (!drv || !n_ents)
>>                 return NULL;
>>
>> -       len = strlen(name);
>> -
>>         for (entry = drv; entry != drv + n_ents; entry++) {
>> -               if (strncmp(name, entry->name, len))
>> -                       continue;
>> -
>> -               /* Full match */
>> -               if (len == strlen(entry->name))
>> +               if (!strcmp(name, entry->name))
>>                         return entry;
>>         }
> 
> The discussion on this was here:
> 
> http://lists.denx.de/pipermail/u-boot/2014-September/189336.html
> 
> I don't see a lot of value in the extra code, so I think we should
> take this patch. If it is found to be a problem, we can go back to the
> defensive code and add a test case so it is clear what exactly we are
> defending against.
> 
> Acked-by: Simon Glass <sjg at chromium.org>
> 

Following the discussion referenced above, we have here a classic
case of C language strings problem.
One can dig about it all over the Internet (for example here [1]).
I don't think we will invent a solution for that problem here.
Also we are not about to abandon the C language...
So, unless someone comes out with a real case to solve, I think
we should merge this.

Acked-by: Igor Grinberg <grinberg at compulab.co.il>

[1] http://www.sei.cmu.edu/library/abstracts/news-at-sei/feature120061.cfm



-- 
Regards,
Igor.