They're nearly identical, so don't use two copies. Merge the newer
driver into the older one, and move it to a common location.
Add the Semihalf and associated copyrights in addition to mine, since
it's a non-trivial amount of code merged.
Details
Details
Diff Detail
Diff Detail
- Repository
- rS FreeBSD src repository - subversion
- Lint
Lint Skipped - Unit
Tests Skipped - Build Status
Buildable 38224 Build 35113: arc lint + arc unit
Event Timeline
Comment Actions
Thank you Justin for doing the change!
| sys/conf/files.arm64 | ||
|---|---|---|
| 106 | It looks like some conflict resolution remainings slipped into the patch. | |
| 216 | Please align 'optional' and the actual options with the entries above. | |
| sys/dev/gpio/qoriq_gpio.c | ||
| 406 | (While not having the docs in hand) sys/arm64/qoriq/ls1046_gpio.c did not use GPIO_GPIBE - this could be a bug, but:
| |
| sys/conf/files.arm64 | ||
|---|---|---|
| 106 | Oops. Must've snuck in during a rebase. | |
| sys/dev/gpio/qoriq_gpio.c | ||
| 406 | This is required for LX2160, but may not be required for others. Every reference I have, for P1022, P5020, and LX2160, any unimplemented registers in this block are ignored, so it's not a problem on other targets. I don't yet have a Honeycomb board, so have only tested on a P1022 board thus far, that it's not broken. This only affects pins marked as input, it has no effect on other pins. This is set 'unconditionally' (only conditional on compatible property) in Linux, too. | |