Re: [Qemu-devel] [PATCHv8 1/3] fw_cfg: switch fw_cfg_find() to locate the fw_cfg device by type rather than path

[Mark Cave-Ayland]

On 11/07/17 21:10, Eduardo Habkost wrote:

> On Tue, Jul 11, 2017 at 09:02:11PM +0100, Mark Cave-Ayland wrote:
>> This will enable the fw_cfg device to be placed anywhere within the QOM tree
>> regardless of its machine location.
>>
>> Note that we also add a comment to document the behaviour that we return 
>> NULL to
>> indicate failure where either no fw_cfg device or multiple fw_cfg devices are
>> found.
>>
>> Signed-off-by: Mark Cave-Ayland <address@hidden>
>> ---
>>  hw/nvram/fw_cfg.c |    7 ++++++-
>>  1 file changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c
>> index 99bdbc2..8ef889a 100644
>> --- a/hw/nvram/fw_cfg.c
>> +++ b/hw/nvram/fw_cfg.c
>> @@ -1017,7 +1017,12 @@ FWCfgState *fw_cfg_init_mem(hwaddr ctl_addr, hwaddr 
>> data_addr)
>>  
>>  FWCfgState *fw_cfg_find(void)
>>  {
>> -    return FW_CFG(object_resolve_path(FW_CFG_PATH, NULL));
>> +    /* Returns FWCfgState if only one fw_cfg device type exists. If zero or
>> +       more than one fw_cfg device are found then NULL is returned as per 
>> the
>> +       object_resolve_path_type() documentation. This behaviour is correct 
>> as
>> +       it ensures that we detect both missing fw_cfg devices and multiple
>> +       fw_cfg devices which could result in unpredictable behaviour. */
>> +    return FW_CFG(object_resolve_path_type("", TYPE_FW_CFG, NULL));
>>  }
> 
> I believe a one-line "returns NULL unless there is exactly one
> fw_cfg device" (similar to the one at find_vmgenid_dev()) would
> suffice.  :)

I got the impression from the thread that Igor thought the semantics of
fw_cfg_find() weren't clear enough, so wanted to make sure a good
explanation was included in the patch.

I'll wait to see if there's any further feedback before a v9...


ATB,

Mark.