hid-tmff2/src/tmt300rs, branch masterACTIVE - Linux driver for Thrustmaster wheelshttps://metanimi.dy.fi/cgit/hid-tmff2/atom?h=master2026-08-08T22:20:18Zhandle full signed effect magnitude range2026-08-08T22:20:18ZKai Krakowkai@kaishome.de2026-08-08T11:04:25Zurn:sha1:1914c478dec801e98e9b73bbfd5bbdc4c0a11cf8
+ Direction scaling can produce 32768 for the valid input pair -32768 and a
negative full-scale sine value. Storing the intermediate result in s16
wraps it before constant, ramp or periodic effects are converted.
+ Keep direction-scaled values in s32 until their destination range is
known. Periodic effects cannot represent a magnitude above S16_MAX, so
clamp the asymmetric endpoint after making the magnitude positive. This
also prevents the following headroom calculation from using a negative
wrapped magnitude.
Revert "use on_hid_hw_open instead of input_dev->open"2026-01-26T16:21:43ZKimplulkimi.h.kuparinen@gmail.com2026-01-26T16:21:43Zurn:sha1:66e522e26549afab26d032e900ae9f6576c83b9d
This reverts commit c26573d77369c042dc2b9a881a7f2ff88ddc0dd9.
+ `on_hid_hw_open` is apparently new enough that common distros don't
have it yet, so let's wait a few years and try again. The old method
works well enough and if not, just set `open_mode=0`.
use on_hid_hw_open instead of input_dev->open2026-01-04T19:49:58ZKimplulkimi.h.kuparinen@gmail.com2026-01-04T19:49:58Zurn:sha1:c26573d77369c042dc2b9a881a7f2ff88ddc0dd9
+ Seems some applications can mess with input_dev->open, still unclear
how but on_hid_hw_open seems to be a bit more reliable so use it for
now at least
fix maybe uninitialized variable2025-12-01T16:56:05ZKimplulkimi.h.kuparinen@gmail.com2025-12-01T16:56:05Zurn:sha1:c4b70d4ec36404c9454199bec208eaf3116ef453
+ Closes #183
minimize critical section in work handler2025-11-03T18:17:32ZKimplulkimi.h.kuparinen@gmail.com2025-11-03T18:17:32Zurn:sha1:5095d47320c36e2346d41eb7e35dcbd215886557
+ Effects handlers were being called in a locked context,
which in this case happened to work due to how `hid_hw_request`
is implemented. Now, only the selected state is updated and the heavy
lifting of making the wheel aware of the requested changes is done
after the critical section.
Merge branch 'refactor-duration' of github.com:MmAaXx500/hid-tmff2 into MmAaXx500-refactor-duration2025-07-28T20:30:41ZKimplulkimi.h.kuparinen@gmail.com2025-07-28T20:30:41Zurn:sha1:303d2d85c471fc2982ca2d15291b06d44d25b425refactor effect length calculation2025-07-24T20:31:59ZMmAaXx500viktor.balogh45@gmail.com2025-07-24T18:19:52Zurn:sha1:d757f6183c650706b5e98d7c55ce7f4b33b5539d
- do the calculation in only one place
- avoid losing 1 ms (if it ever matters)
fix periodic effect length on update2025-07-24T17:14:55ZMmAaXx500viktor.balogh45@gmail.com2025-07-24T16:39:38Zurn:sha1:7388235046b02cff0921fbbfd2b9843a935476fb
Avoid subtracting twice from the length of the effect
first time in t300rs_calculate_periodic_values
second time in t300rs_update_periodic
make firmware check a warning2024-12-18T20:06:18ZKimplulkimi.h.kuparinen@gmail.com2024-12-18T20:06:18Zurn:sha1:9814765860ed49e482481f30ee3fd73102348951
+ Apparently this allows the T-GT II to work, as it shows up as a PS4
T300 RS, just that it has a different firmware version.
Eventually I should maybe look into the firmware response a bit closer
and check if I can fish out wheel information or something along those
lines to improve wheel detection.
fix style2024-09-01T15:32:39ZMmAaXx500viktor.balogh45@gmail.com2024-09-01T14:41:29Zurn:sha1:901b37cd82a002da9112fa2882eb2fdfa2f6729b