Re: [Qemu-devel] [ARM SMBIOS V4 PATCH 1/2] smbios: add smbios 3.0 support

[Wei Huang]

On 08/25/2015 10:29 AM, Leif Lindholm wrote:
> On Tue, Aug 25, 2015 at 04:17:42PM +0100, Peter Maydell wrote:
>> On 13 August 2015 at 18:09, Wei Huang <address@hidden> wrote:
>>> This patch adds support for SMBIOS 3.0 entry point. When caller invokes
>>> smbios_set_defaults(), it can specify entry point as 2.1 or 3.0. Then
>>> smbios_get_tables() will return the entry point table in right format.
>>
>>
>>> -/* SMBIOS entry point (anchor).
>>> - * BIOS must place this at a 16-bit-aligned address between 0xf0000 and 
>>> 0xfffff.
>>> +/* SMBIOS entry point
>>> + * BIOS must place this at a 16-bit-aligned address between 0xf0000
>>> + * and 0xfffff.
>>>   */
>>> -struct smbios_entry_point {
>>> +struct smbios_21_entry_point {
>>>      uint8_t anchor_string[4];
>>>      uint8_t checksum;
>>>      uint8_t length;
>>> @@ -58,6 +52,25 @@ struct smbios_entry_point {
>>>      uint8_t smbios_bcd_revision;
>>>  } QEMU_PACKED;
>>
>> This breaks 'make check' for x86, because tests/bios-tables-test.c
>> still uses 'struct smbios_entry_point' and no longer compiles
>> if this patch is applied.
> 
> Urgh.
> 
>> I'm removing these two patches from my target-arm queue.
> 
> Fair enough.
> 
> Wei - is there actually any particular point in renaming this
> structure? In all versions of the specification before 3.0, this was
> only known as the "smbios entry point". Only with the introduction of
> SMBIOS 3.0 this was retrospectively renamed.

I can take this suggestion, with clear comment in header file so nobody
will get confused. Peter, please let me know if you object.

The new patches should be out in a short while.

Thanks,
-Wei

> 
> (And personally, I find that renaming a bit counterintuitive, since it
> is still a valid 32-bit entry point in SMBIOS3, and was the only entry
> point up until and including SMBIOS 2.8.)


> 
> /
>     Leif
>