Use a specific device ?

[Jean-Christian de Rivaz]

Le 10. 06. 15 23:37, Bjørn Mork a écrit :
> Jean-Christian de Rivaz <jc at eclis.ch> writes:
>
>> There is not so
>> much modem manufacturers and each of them don't even release a new
>> product range per year.
> Ehh... I don't think we live on the same planet.  Did you know Toshiba
> is a "modem manufacturer"? Dell? HP? There are 43 (damn - I would have
> loved to see 42) different vendor IDs just in the option driver:
>
>   bjorn at nemi:/usr/local/src/git/linux$ git grep -E '^#define.*VENDOR' drivers/usb/serial/option.c |wc -l
>   43

Please provide a complete picture:
git grep -E '^#define.*VENDOR' drivers/usb/serial/* | wc -l
174

Not a such bit number. There are various vendor/product database on the 
internet, I failed to identify a unmanageable number of modem on them.

> Feel free to start updating the whitelists in vendor specific drivers
> like option, qcserial and qmi_wwan. Please let me know when those are
> complete.
>
> No, I don't seriously expect you to do that job.  Fact is that the
> whitelists are unmaintablable even when the scope is limited to one
> specific mode of Qualcomm based modems.  Keeping a semi-complete
> whitelist of all modems is not going to happen.
>

I expected this reaction. The first problem was the fact that 
ModemManager is unable to provide a stable name to NetworkManager 
requiring a hack in the configuration that will not work anymore if 
there more than a single modem. Instead of making some constructives 
propositions to find a way to sole the problem clearly caused by 
ModemManager, the project push all his effort to reject any critic and 
constructive proposition. I proposed a way to make the transition to a 
whitelist but you don't even comment on it: you seem only focused on 
rejecting anything that can possibly fix the ModemManager problems.

>> The 40-usb_modeswitch.rules required by some
>> modems is not so big either.
> There are approximately a gazillion modem IDs which do *not* need mode
> switching.  But list size is irrelevant in any case.  See below.

So why did the list exists?

>> But most important is to understand that the current ModemManager is
>> abusing the udev concept and confusing the users. Are you really
>> serious when you ask a random people with a new UPS product to add a
>> new udev rule to the ModemManager project?
> Why do they have to do that? Their UPS should work fine even if MM
> happens to probe it.

So why did you need blacklist in the first place? The only fact that the 
blacklist exists is a prove that users don't want probing for non modem 
device. The UPS is maybe not the best example, but you can't deny the 
reality of the blacklist.

>> I think you are so focused
>> on defending the current ModemManager abomination that you fail to see
>> the problem from the point of view of a common user.
> So, let's try to agree what a common user wants.  My claim is that the
> common user wants *both* their UPS and 3G modem to work by default.
>
>> The length of a
>> white list is not an excuse to not fix the problem.
> Agreed.  The length is irrelevant. The problem is that the list will be
> incomplete, whether it is a blacklist or whitelist. We could probably
> discuss which one will be easiest to maintain, but that's really
> irrelevant too. The real question is what happens to the user
> requirements in the two cases, assuming that we don't have any prior
> knowledge of the devices (which is very likely for any device you can
> buy new in a store):
>
> whitelist:
>   UPS works
>   3G modem fails
>
> blacklist:
>   UPS works, but is unnecessarily probed by MM
>   3G modem works
>
> The choice is really simple, isn't it? MM does what it has to do.
>

Now you have completely changed the situation that cause problem to a 
situation where you don't see a problem. You may be happy doing this, 
but the former problem is still there and unsolved.

Jean-Christian