HI, Am 22.10.2015 um 00:40 schrieb Denis 'GNUtoo' Carikli <[email protected]>:
> On Thu, 3 Sep 2015 15:38:33 +0200 > "Dr. H. Nikolaus Schaller" <[email protected]> wrote: > >> e.g. OPTION modules have a HSO driver in Linux. Gemalto modules do >> not even need a driver because they use the standard ttyACM protocol >> and PPP. >> >> So I am not exactly clear which problem you want to solve. It is IMHO >> not availability of such isolated solutions but that smart phone >> vendors don’t use them (for space and cost reasons). > The Linux drivers should work out of the box, but they are only half of > the solution. > Now as Paul pointed out, some need specific userspace protocol > handlers. > I've no experience with sourcing modems, but if you buy modem modules, > how can you know in advance that it will work well, for a smartphone > usage? study the data sheet. It explains everything. Or if there is no or no sufficient data sheet available (even if under NDA), don't buy it... > More specifically, how to you reduce the risk of having nasty > undocumented bugs, with undocumented workarounds that the existing free > software isn't aware of? You always have to be aware of bugs and some are indeed nasty... It is the same with e.g. CPUs. If you are lucky, there is an errata document, but then they are documented. IMHO, undocumented bugs are not specific to modems. A different thing is how difficult it is to get documentation. It may not be available to the public, but it may be available. And how helpful the documentation is to fix a bug. For example we do have the OMAP3 TRM (5000 pages PDF) but it does not yet directly help to find the energy hog not turned off by the GTA04 Kernel. > I vaguely remember the GTA04 having similar issues, and I was impressed > at how fast it was fixed by the community. I don't remember if the AT > commands that fixed it appeared inside the list of supported commands, > or if it was documented elsewhere already. A mix of all of this. BR, NIkolaus _______________________________________________ Replicant mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/replicant
