annotate doc/Modem-operation @ 275:79cfefc1e2b4

audio mode load: gracefully handle mode files of wrong AEC version Unfortunately our change of enabling L1_NEW_AEC (which is necessary in order to bring our Calypso ARM fw into match with the underlying DSP reality) brings along a change in the audio mode file binary format and file size - all those new tunable AEC parameters do need to be stored somewhere, after all. But we already have existing mode files in the old format, and setting AEC config to garbage when loading old audio modes (which is what would happen without the present change) is not an appealing proposition. The solution implemented in the present change is as follows: the audio mode loading code checks the file size, and if it differs from the active version of T_AUDIO_MODE, the T_AUDIO_AEC_CFG structure is cleared - set to the default (disabled AEC) for the compiled type of AEC. We got lucky in that this varying T_AUDIO_AEC_CFG structure sits at the end of T_AUDIO_MODE!
author Mychaela Falconia <falcon@freecalypso.org>
date Fri, 30 Jul 2021 02:55:48 +0000
parents a62e5bf88434
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
50
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
1 FC Tourmaline stdmodem functional configuration is equivalent to Magnetite
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
2 hybrid. The only intentional diff from Magnetite to Tourmaline in the modem
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
3 configuration is the change of Nucleus: Magnetite uses the blob version of
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
4 Nucleus which came with TCS211-20070608, whereas Tourmaline uses the source
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
5 version of Nucleus by Comrade XVilka. It is essentially a change from an
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
6 unknown version of Nucleus to a known one, thus it should generally be
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
7 considered an improvement - but as always with such changes, extensive
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
8 verification and testing needs to be done in order to ensure that nothing got
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
9 broken.
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
10
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
11 Aside from the one intentional change of Nucleus version, FC Tourmaline is an
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
12 entirely new firmware source tree, thus it is always possible that something
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
13 could have got broken unintentionally, particularly in rarely-tested areas of
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
14 the firmware. Therefore, extensive testing will need to be done before we can
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
15 declare Tourmaline stdmodem as officially replacing Magnetite hybrid for stable
a62e5bf88434 first round of documentation
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff changeset
16 modem products.