| 29 May 2025 |
 matthewcroughan | I'm not sure under what circumstances this bin file will be used by this name | 12:33:29 |
| matthewcroughan | static char *mt7996_eeprom_name(struct mt7996_dev *dev)
{
switch (mt76_chip(&dev->mt76)) {
case MT7992_DEVICE_ID:
switch (dev->var.type) {
case MT7992_VAR_TYPE_23:
if (dev->var.fem == MT7996_FEM_INT)
return MT7992_EEPROM_DEFAULT_23_INT;
return MT7992_EEPROM_DEFAULT_23;
case MT7992_VAR_TYPE_44:
default:
if (dev->var.fem == MT7996_FEM_INT)
return MT7992_EEPROM_DEFAULT_INT;
if (dev->var.fem == MT7996_FEM_MIX)
return MT7992_EEPROM_DEFAULT_MIX;
return MT7992_EEPROM_DEFAULT;
}
case MT7990_DEVICE_ID:
if (dev->var.fem == MT7996_FEM_INT)
return MT7990_EEPROM_DEFAULT_INT;
return MT7990_EEPROM_DEFAULT;
case MT7996_DEVICE_ID:
default:
switch (dev->var.type) {
case MT7996_VAR_TYPE_233:
if (dev->var.fem == MT7996_FEM_INT)
return MT7996_EEPROM_DEFAULT_233_INT;
return MT7996_EEPROM_DEFAULT_233;
case MT7996_VAR_TYPE_444:
default:
if (dev->var.fem == MT7996_FEM_INT)
return MT7996_EEPROM_DEFAULT_INT;
return MT7996_EEPROM_DEFAULT;
}
}
}
| 12:33:47 |
| matthewcroughan | I guess it just uses this if it's 7996 at all | 12:34:28 |
| matthewcroughan | Their google drive filename is interesting | 12:35:08 |
| matthewcroughan | Eagle_MT7976C_MT7977iA_2i3i3i_6G_TSSI_OFF_TC_ON_Test_mode_20240116_V04.bin | 12:35:11 |
 emily | my guess from glancing very briefly at code is that it is unconditionally used as firmware, but the txpower calibration data from the device is preferred over the defaults in the firmware. | 12:35:17 |
| matthewcroughan | So they have access to the board files in Eagle I guess | 12:35:21 |
| emily | (for obvious reasons, but that's what causes the issue when your data is busted) | 12:35:27 |
| matthewcroughan | And I guess those bytes you're seeing as turning TSSI off and TC on | 12:35:36 |
| emily | isn't Eagle the codename of the Filogic stuff? | 12:35:45 |
| matthewcroughan | and enabling test mode? | 12:35:45 |
| matthewcroughan | may be, but it's also a cad package isn't it | 12:36:02 |
| emily | is that based on the offsets the kernel driver loks at? | 12:36:06 |
| emily | * is that based on the offsets the kernel driver looks at? | 12:36:07 |
| emily | oh | 12:36:11 |
| emily | it's based on the filename | 12:36:15 |
| matthewcroughan | I can try this firmware alongside the txpower fix to see if there's a difference | 12:36:21 |
| emily | I do not think we can conclude that – it may be that that's the specific configuration they pick regardless of version. | 12:36:25 |
| emily | i.e., Filogic sends them a bunch of firmware with various configurations and they picked that one. | 12:37:12 |
| emily | MediaTek rather. whatever. | 12:37:19 |
| matthewcroughan | I tried to get WED working quite unsuccessfully btw | 12:37:37 |
| matthewcroughan | mt7996-firmware = final.runCommand "mt7996-firmware" { } ''
mkdir -p $out/lib/firmware/mediatek
cp -rv ${final.bpi-r4.mt76} $out/lib/firmware/mediatek/mt7996
cp -rv ${./mt7988_wo_0.bin} $out/lib/firmware/mediatek/mt7988_wo_0.bin
cp -rv ${./mt7988_wo_1.bin} $out/lib/firmware/mediatek/mt7988_wo_1.bin
cp -rv ${./mt7988_wo_2.bin} $out/lib/firmware/mediatek/mt7988_wo_2.bin
ls -lah $out/lib/firmware/mediatek/mt7996
'';
| 12:37:45 |
| matthewcroughan | mt7996-firmware = final.runCommand "mt7996-firmware" { } ''
mkdir -p $out/lib/firmware/mediatek
cp -rv ${final.bpi-r4.mt76} $out/lib/firmware/mediatek/mt7996
cp -rv ${./mt7988_wo_0.bin} $out/lib/firmware/mediatek/mt7988_wo_0.bin
cp -rv ${./mt7988_wo_1.bin} $out/lib/firmware/mediatek/mt7988_wo_1.bin
cp -rv ${./mt7988_wo_2.bin} $out/lib/firmware/mediatek/mt7988_wo_2.bin
ls -lah $out/lib/firmware/mediatek/mt7996
';
| 12:37:47 |
| matthewcroughan | Redacted or Malformed Event | 12:38:09 |
| matthewcroughan | Redacted or Malformed Event | 12:38:21 |
| matthewcroughan | Redacted or Malformed Event | 12:38:47 |
| matthewcroughan | Redacted or Malformed Event | 12:41:20 |
| matthewcroughan | Redacted or Malformed Event | 13:04:17 |
| emily | this is happening again :/ | 13:10:20 |
| emily | a few minutes after some
[53464.014645] Virtual device eth0 asks to queue packet!
[54556.087872] Virtual device eth0 asks to queue packet!
[54556.092961] Virtual device eth0 asks to queue packet!
[54782.344423] Virtual device eth0 asks to queue packet!
| 13:10:39 |
