Bug 27481
| Summary: | [PATCH] to make snp(4) a module | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Base System | Reporter: | dd <dd> | ||||
| Component: | kern | Assignee: | dd <dd> | ||||
| Status: | Closed FIXED | ||||||
| Severity: | Affects Only Me | ||||||
| Priority: | Normal | ||||||
| Version: | Unspecified | ||||||
| Hardware: | Any | ||||||
| OS: | Any | ||||||
| Attachments: |
|
||||||
Responsible Changed From-To: freebsd-bugs->dd Dima should just commit this - maybe with a lines saying ``Reviewed by: silence on -arch and silence on -committers''. State Changed From-To: open->closed Committed my own patch. |
Right now, snp(4) can't be compiled as a module because it depends on hacks in the tty subsystem similar to the following: #ifdef DEV_SNP if (ISSET(tp->t_state, TS_SNOOP) && tp->t_sc != NULL) snpin((struct snoop *)tp->t_sc, cp, cc); #endif snpin() is a function in snp(4), so it's not simply a matter of removing the #ifdef's. Attached is a patch that converts snp(4) to use line disciplines to get the information it needs. I'm not sure if it's entirely appropriate to use line disciplines this way, but I think it's a lot better than the above code, and it has the fortunate sideaffect that snp(4) can be built as a module. If this gets accepted I'll submit patches to remove the aforementioned tty hacks. Fix: I've sent this to both -hackers and -audit; nobody replied. I take this to mean that nobody cares, and my use of line disciplines isn't evil enough to warrant attention. If somebody just wants to review this so I can commit it (I'm not a src/ committer) that'd be fine, too.