linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 1/1] DMI: fix dmi_get_year year parsing
@ 2009-07-07 20:36 Jiri Slaby
  2009-07-08  6:46 ` Matthias Pfaller
  0 siblings, 1 reply; 6+ messages in thread
From: Jiri Slaby @ 2009-07-07 20:36 UTC (permalink / raw)
  To: akpm; +Cc: leo, linux-kernel, Jiri Slaby

Don't guess a year number base. Use 10 instead, since year may
be 2-digit starting with 0, so that we would end up in base equal
to 8.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Reported-by: Matthias Pfaller <leo@marco.de>
---
 drivers/firmware/dmi_scan.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
index 6071078..8fe0f6e 100644
--- a/drivers/firmware/dmi_scan.c
+++ b/drivers/firmware/dmi_scan.c
@@ -611,7 +611,7 @@ int dmi_get_year(int field)
 		return 0;
 
 	s += 1;
-	year = simple_strtoul(s, NULL, 0);
+	year = simple_strtoul(s, NULL, 10);
 	if (year && year < 100) {	/* 2-digit year */
 		year += 1900;
 		if (year < 1996)	/* no dates < spec 1.0 */
-- 
1.6.3.2


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH 1/1] DMI: fix dmi_get_year year parsing
  2009-07-07 20:36 [PATCH 1/1] DMI: fix dmi_get_year year parsing Jiri Slaby
@ 2009-07-08  6:46 ` Matthias Pfaller
  2009-07-08  6:54   ` Jiri Slaby
  0 siblings, 1 reply; 6+ messages in thread
From: Matthias Pfaller @ 2009-07-08  6:46 UTC (permalink / raw)
  To: Jiri Slaby; +Cc: akpm, linux-kernel

Jiri Slaby wrote:
> Don't guess a year number base. Use 10 instead, since year may
> be 2-digit starting with 0, so that we would end up in base equal
> to 8.
> 
> Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
> Reported-by: Matthias Pfaller <leo@marco.de>
> ---
>  drivers/firmware/dmi_scan.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
> index 6071078..8fe0f6e 100644
> --- a/drivers/firmware/dmi_scan.c
> +++ b/drivers/firmware/dmi_scan.c
> @@ -611,7 +611,7 @@ int dmi_get_year(int field)
>  		return 0;
>  
>  	s += 1;
> -	year = simple_strtoul(s, NULL, 0);
> +	year = simple_strtoul(s, NULL, 10);
>  	if (year && year < 100) {	/* 2-digit year */
>  		year += 1900;
>  		if (year < 1996)	/* no dates < spec 1.0 */

I just noticed, that this is not enough, because this will still fail 
for xxx/xx/00. I suggest the following patch:

--- drivers/firmware/dmi_scan.c.bak     Wed Jul  8 02:42:04 2009
+++ drivers/firmware/dmi_scan.c Wed Jul  8 02:42:17 2009
@@ -360,12 +360,15 @@
                 return 0;

         s += 1;
-       year = simple_strtoul(s, NULL, 0);
-       if (year && year < 100) {       /* 2-digit year */
-               year += 1900;
-               if (year < 1996)        /* no dates < spec 1.0 */
-                       year += 100;
+       if (s[0] == '0' && s[1] == '0' && s[2] == '\0') {
+               year = 2000;
+       } else {
+               year = simple_strtoul(s, NULL, 10);
+               if (year && year < 100) {       /* 2-digit year */
+                       year += 1900;
+                       if (year < 1996)        /* no dates < spec 1.0 */
+                               year += 100;
+               }
         }
-
         return year;
  }

-- 
Matthias Pfaller                          Software Entwicklung
marco Systemanalyse und Entwicklung GmbH  Tel   +49 8131 5161 41
Hans-Böckler-Str. 2, D 85221 Dachau       Fax   +49 8131 5161 66
http://www.marco.de/                      Email leo@marco.de
Geschäftsführer Martin Reuter             HRB 171775 Amtsgericht München

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 1/1] DMI: fix dmi_get_year year parsing
  2009-07-08  6:46 ` Matthias Pfaller
@ 2009-07-08  6:54   ` Jiri Slaby
  2009-07-08  7:03     ` Matthias Pfaller
  0 siblings, 1 reply; 6+ messages in thread
From: Jiri Slaby @ 2009-07-08  6:54 UTC (permalink / raw)
  To: Matthias Pfaller; +Cc: akpm, linux-kernel, Eric W. Biederman

On 07/08/2009 08:46 AM, Matthias Pfaller wrote:
> Jiri Slaby wrote:
>> Don't guess a year number base. Use 10 instead, since year may
>> be 2-digit starting with 0, so that we would end up in base equal
>> to 8.
>>
>> Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
>> Reported-by: Matthias Pfaller <leo@marco.de>
>> ---
>>  drivers/firmware/dmi_scan.c |    2 +-
>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
>> index 6071078..8fe0f6e 100644
>> --- a/drivers/firmware/dmi_scan.c
>> +++ b/drivers/firmware/dmi_scan.c
>> @@ -611,7 +611,7 @@ int dmi_get_year(int field)
>>          return 0;
>>  
>>      s += 1;
>> -    year = simple_strtoul(s, NULL, 0);
>> +    year = simple_strtoul(s, NULL, 10);
>>      if (year && year < 100) {    /* 2-digit year */
>>          year += 1900;
>>          if (year < 1996)    /* no dates < spec 1.0 */
> 
> I just noticed, that this is not enough, because this will still fail
> for xxx/xx/00. I suggest the following patch:

Actually the patch below is not correct. Standard says consider xx/xx/yy
as 19yy, not 20yy.

BTW. the patch above is not useful, reporting 1908 and 1909 as a year is
almost the same as 1900.

Do you have such a broken BIOS?

> --- drivers/firmware/dmi_scan.c.bak     Wed Jul  8 02:42:04 2009
> +++ drivers/firmware/dmi_scan.c Wed Jul  8 02:42:17 2009
> @@ -360,12 +360,15 @@
>                 return 0;
> 
>         s += 1;
> -       year = simple_strtoul(s, NULL, 0);
> -       if (year && year < 100) {       /* 2-digit year */
> -               year += 1900;
> -               if (year < 1996)        /* no dates < spec 1.0 */
> -                       year += 100;
> +       if (s[0] == '0' && s[1] == '0' && s[2] == '\0') {
> +               year = 2000;
> +       } else {
> +               year = simple_strtoul(s, NULL, 10);
> +               if (year && year < 100) {       /* 2-digit year */
> +                       year += 1900;
> +                       if (year < 1996)        /* no dates < spec 1.0 */
> +                               year += 100;
> +               }
>         }
> -
>         return year;
>  }
> 


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 1/1] DMI: fix dmi_get_year year parsing
  2009-07-08  6:54   ` Jiri Slaby
@ 2009-07-08  7:03     ` Matthias Pfaller
  2009-07-09  7:12       ` DMI year "00" + ACPI [was: DMI: fix dmi_get_year year parsing] Jiri Slaby
  0 siblings, 1 reply; 6+ messages in thread
From: Matthias Pfaller @ 2009-07-08  7:03 UTC (permalink / raw)
  To: Jiri Slaby, akpm, linux-kernel, Eric W. Biederman

Jiri Slaby wrote:
> On 07/08/2009 08:46 AM, Matthias Pfaller wrote:
>> Jiri Slaby wrote:
>>> Don't guess a year number base. Use 10 instead, since year may
>>> be 2-digit starting with 0, so that we would end up in base equal
>>> to 8.
>>>
>>> Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
>>> Reported-by: Matthias Pfaller <leo@marco.de>
>>> ---
>>>  drivers/firmware/dmi_scan.c |    2 +-
>>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>>
>>> diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
>>> index 6071078..8fe0f6e 100644
>>> --- a/drivers/firmware/dmi_scan.c
>>> +++ b/drivers/firmware/dmi_scan.c
>>> @@ -611,7 +611,7 @@ int dmi_get_year(int field)
>>>          return 0;
>>>  
>>>      s += 1;
>>> -    year = simple_strtoul(s, NULL, 0);
>>> +    year = simple_strtoul(s, NULL, 10);
>>>      if (year && year < 100) {    /* 2-digit year */
>>>          year += 1900;
>>>          if (year < 1996)    /* no dates < spec 1.0 */
>> I just noticed, that this is not enough, because this will still fail
>> for xxx/xx/00. I suggest the following patch:
> 
> Actually the patch below is not correct. Standard says consider xx/xx/yy
> as 19yy, not 20yy.
> 
> BTW. the patch above is not useful, reporting 1908 and 1909 as a year is
> almost the same as 1900.

Sorry, I don't understand. simple_strtoul will return 8 for my bios. So
1900 will get added resulting in 1908 (as you said). But because 1908 is 
still less than 1996 now 100 will get added which results in 2008 which 
is what I need.

> Do you have such a broken BIOS?

Yes. Its on an etx-cd, an embedded board from kontron. dmidecode output:

Handle 0x0000, DMI type 0, 24 bytes
BIOS Information
         Vendor: Phoenix Technologies LTD
         Version: MCALR116
         Release Date: 08/13/08
         Address: 0xE3F90
         Runtime Size: 114800 bytes
         ROM Size: 1024 kB
         Characteristics:
                 ISA is supported
                 PCI is supported
                 PC Card (PCMCIA) is supported
                 PNP is supported
                 BIOS is upgradeable
                 BIOS shadowing is allowed
                 ESCD support is available
                 Boot from CD is supported
                 ACPI is supported
                 USB legacy is supported
                 BIOS boot specification is supported

Our systems started to fail after we got boards with a newer bios from 
our supplier...

> 
>> --- drivers/firmware/dmi_scan.c.bak     Wed Jul  8 02:42:04 2009
>> +++ drivers/firmware/dmi_scan.c Wed Jul  8 02:42:17 2009
>> @@ -360,12 +360,15 @@
>>                 return 0;
>>
>>         s += 1;
>> -       year = simple_strtoul(s, NULL, 0);
>> -       if (year && year < 100) {       /* 2-digit year */
>> -               year += 1900;
>> -               if (year < 1996)        /* no dates < spec 1.0 */
>> -                       year += 100;
>> +       if (s[0] == '0' && s[1] == '0' && s[2] == '\0') {
>> +               year = 2000;
>> +       } else {
>> +               year = simple_strtoul(s, NULL, 10);
>> +               if (year && year < 100) {       /* 2-digit year */
>> +                       year += 1900;
>> +                       if (year < 1996)        /* no dates < spec 1.0 */
>> +                               year += 100;
>> +               }
>>         }
>> -
>>         return year;
>>  }
>>
> 


-- 
Matthias Pfaller                          Software Entwicklung
marco Systemanalyse und Entwicklung GmbH  Tel   +49 8131 5161 41
Hans-Böckler-Str. 2, D 85221 Dachau       Fax   +49 8131 5161 66
http://www.marco.de/                      Email leo@marco.de
Geschäftsführer Martin Reuter             HRB 171775 Amtsgericht München

^ permalink raw reply	[flat|nested] 6+ messages in thread

* DMI year "00" + ACPI [was: DMI: fix dmi_get_year year parsing]
  2009-07-08  7:03     ` Matthias Pfaller
@ 2009-07-09  7:12       ` Jiri Slaby
  2009-07-09  7:39         ` Matthias Pfaller
  0 siblings, 1 reply; 6+ messages in thread
From: Jiri Slaby @ 2009-07-09  7:12 UTC (permalink / raw)
  To: Matthias Pfaller
  Cc: akpm, linux-kernel, Eric W. Biederman, Len Brown, linux-acpi

On 07/08/2009 09:03 AM, Matthias Pfaller wrote:
> Jiri Slaby wrote:
>> On 07/08/2009 08:46 AM, Matthias Pfaller wrote:
>>> Jiri Slaby wrote:
>>>> Don't guess a year number base. Use 10 instead, since year may
>>>> be 2-digit starting with 0, so that we would end up in base equal
>>>> to 8.
>>>>
>>>> Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
>>>> Reported-by: Matthias Pfaller <leo@marco.de>
>>>> ---
>>>>  drivers/firmware/dmi_scan.c |    2 +-
>>>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>>>
>>>> diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
>>>> index 6071078..8fe0f6e 100644
>>>> --- a/drivers/firmware/dmi_scan.c
>>>> +++ b/drivers/firmware/dmi_scan.c
>>>> @@ -611,7 +611,7 @@ int dmi_get_year(int field)
>>>>          return 0;
>>>>  
>>>>      s += 1;
>>>> -    year = simple_strtoul(s, NULL, 0);
>>>> +    year = simple_strtoul(s, NULL, 10);
>>>>      if (year && year < 100) {    /* 2-digit year */
>>>>          year += 1900;
>>>>          if (year < 1996)    /* no dates < spec 1.0 */
>>> I just noticed, that this is not enough, because this will still fail
>>> for xxx/xx/00. I suggest the following patch:
>>
>> Actually the patch below is not correct. Standard says consider xx/xx/yy
>> as 19yy, not 20yy.
>>
>> BTW. the patch above is not useful, reporting 1908 and 1909 as a year is
>> almost the same as 1900.
> 
> Sorry, I don't understand.

Nevermind, you're right. Both patches are OK, however we might hit a
regression with 00/00/00 entries so that acpi gets unintentionally armed
now for them (ACPI checks year == 0). I don't know if it's worth it. Any
ideas?

>>> --- drivers/firmware/dmi_scan.c.bak     Wed Jul  8 02:42:04 2009
>>> +++ drivers/firmware/dmi_scan.c Wed Jul  8 02:42:17 2009
>>> @@ -360,12 +360,15 @@
>>>                 return 0;
>>>
>>>         s += 1;
>>> -       year = simple_strtoul(s, NULL, 0);
>>> -       if (year && year < 100) {       /* 2-digit year */
>>> -               year += 1900;
>>> -               if (year < 1996)        /* no dates < spec 1.0 */
>>> -                       year += 100;
>>> +       if (s[0] == '0' && s[1] == '0' && s[2] == '\0') {
>>> +               year = 2000;
>>> +       } else {
>>> +               year = simple_strtoul(s, NULL, 10);
>>> +               if (year && year < 100) {       /* 2-digit year */
>>> +                       year += 1900;
>>> +                       if (year < 1996)        /* no dates < spec
>>> 1.0 */
>>> +                               year += 100;
>>> +               }
>>>         }
>>> -
>>>         return year;
>>>  }


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: DMI year "00" + ACPI [was: DMI: fix dmi_get_year year parsing]
  2009-07-09  7:12       ` DMI year "00" + ACPI [was: DMI: fix dmi_get_year year parsing] Jiri Slaby
@ 2009-07-09  7:39         ` Matthias Pfaller
  0 siblings, 0 replies; 6+ messages in thread
From: Matthias Pfaller @ 2009-07-09  7:39 UTC (permalink / raw)
  To: Jiri Slaby; +Cc: akpm, linux-kernel, Eric W. Biederman, Len Brown, linux-acpi

Jiri Slaby wrote:
> On 07/08/2009 09:03 AM, Matthias Pfaller wrote:
>> Jiri Slaby wrote:
>>> On 07/08/2009 08:46 AM, Matthias Pfaller wrote:
>>>> Jiri Slaby wrote:
>>>>> Don't guess a year number base. Use 10 instead, since year may
>>>>> be 2-digit starting with 0, so that we would end up in base equal
>>>>> to 8.
>>>>>
>>>>> Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
>>>>> Reported-by: Matthias Pfaller <leo@marco.de>
>>>>> ---
>>>>>  drivers/firmware/dmi_scan.c |    2 +-
>>>>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>>>>
>>>>> diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
>>>>> index 6071078..8fe0f6e 100644
>>>>> --- a/drivers/firmware/dmi_scan.c
>>>>> +++ b/drivers/firmware/dmi_scan.c
>>>>> @@ -611,7 +611,7 @@ int dmi_get_year(int field)
>>>>>          return 0;
>>>>>  
>>>>>      s += 1;
>>>>> -    year = simple_strtoul(s, NULL, 0);
>>>>> +    year = simple_strtoul(s, NULL, 10);
>>>>>      if (year && year < 100) {    /* 2-digit year */
>>>>>          year += 1900;
>>>>>          if (year < 1996)    /* no dates < spec 1.0 */
>>>> I just noticed, that this is not enough, because this will still fail
>>>> for xxx/xx/00. I suggest the following patch:
>>> Actually the patch below is not correct. Standard says consider xx/xx/yy
>>> as 19yy, not 20yy.
>>>
>>> BTW. the patch above is not useful, reporting 1908 and 1909 as a year is
>>> almost the same as 1900.
>> Sorry, I don't understand.
> 
> Nevermind, you're right. Both patches are OK, however we might hit a
> regression with 00/00/00 entries so that acpi gets unintentionally armed
> now for them (ACPI checks year == 0). I don't know if it's worth it. Any
> ideas?

There are bioses with 00/00/00? How about that:

--- dmi_scan.c.orig     Thu Jul  9 03:34:48 2009
+++ dmi_scan.c  Thu Jul  9 03:38:19 2009
@@ -352,17 +352,36 @@
                 return -1;
         if (*s == '\0')
                 return 0;
+
+       /* Month must not be zero */
+       if (simple_strtoul(s, NULL, 10) == 0)
+               return 0;
+
+       /* Look for day of month */
+       s = strchr(s, '/');
+       if (!s)
+               return 0;
+
+       /* Day of month must not be zero */
+       s += 1;
+       if (simple_strtoul(s, NULL, 10) == 0)
+               return 0;
+
+       /* Look for year */
         s = strrchr(s, '/');
         if (!s)
                 return 0;
          s += 1;
-       year = simple_strtoul(s, NULL, 0);
-       if (year && year < 100) {       /* 2-digit year */
-               year += 1900;
-               if (year < 1996)        /* no dates < spec 1.0 */
-                       year += 100;
+       if (s[0] == '0' && s[1] == '0' && s[2] == '\0') {
+               year = 2000;
+       } else {
+               year = simple_strtoul(s, NULL, 10);
+               if (year && year < 100) {       /* 2-digit year */
+                       year += 1900;
+                       if (year < 1996)        /* no dates < spec 1.0 */
+                               year += 100;
+               }
         }
-
         return year;
  }
-- 
Matthias Pfaller                          Software Entwicklung
marco Systemanalyse und Entwicklung GmbH  Tel   +49 8131 5161 41
Hans-Böckler-Str. 2, D 85221 Dachau       Fax   +49 8131 5161 66
http://www.marco.de/                      Email leo@marco.de
Geschäftsführer Martin Reuter             HRB 171775 Amtsgericht München

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2009-07-09  7:46 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-07-07 20:36 [PATCH 1/1] DMI: fix dmi_get_year year parsing Jiri Slaby
2009-07-08  6:46 ` Matthias Pfaller
2009-07-08  6:54   ` Jiri Slaby
2009-07-08  7:03     ` Matthias Pfaller
2009-07-09  7:12       ` DMI year "00" + ACPI [was: DMI: fix dmi_get_year year parsing] Jiri Slaby
2009-07-09  7:39         ` Matthias Pfaller

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).