Details
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Skipped - Unit
Tests Skipped - Build Status
Buildable 55487 Build 52376: arc lint + arc unit
Event Timeline
| sys/dev/mpr/mpr_config.c | ||
|---|---|---|
| 737 | I'd be tempted to add a comment /* already le */ here and maybe elsewhere. It gets confusing why some are swapped and others aren't otherwise. | |
Id also be tempted to always do the requests in host endian and do the happing before we submit. Likewise with the replies. Not sure how feasible that is though.
I think this is the last place to do the swap before the request gets submitted...at least the way it is written right now.
Should mps(4) be updated with this as well? it looks like it has the same issue
| sys/dev/mpr/mpr_config.c | ||
|---|---|---|
| 737 | yea, I know what you mean, especially the confusing part. I think it's slightly confusing since there is some code duplication going on which could be reduced by converting the functions that read/write config pages to use mpr_read_config_page() and mpr_write_config_page(). Then, at least, there would be only a single set of comments explaining that replies from the controller are little-endian. | |
It is grouped in with some other small fixes that I will start submitting reviews for in the coming week.
I was under the impression that mps(4) didn't work at all on big endian. If that's not the case, it too needs the same treatement.
I'd also look at mpi3mr, new to the tree, to see what it needs in this area since it is also a MPT driver