Diff Detail
- Lint
Lint Skipped - Unit
Tests Skipped
Event Timeline
Mostly minor grammar nits in the man page, I'm afraid.
I was a bit concerned that the names of the various constants are confusing, but I'm having trouble thinking of better ones. The best I can come up with is that PROC_TRACE_ENABLE should be PROC_TRACE_SET to match the other constants .. which I think might be better.
| lib/libc/sys/procctl.2 | ||
|---|---|---|
| 284 | Should we mention DTrace here as well? | |
| 293 | Minor grammar nit: don't need 'the' here. | |
| 298 | Similar nit: don't need 'The' here. | |
| 302 | Probably good to say 'on .Xr fork 2' to make it clear what parent-child relationship is meant. Some readers might be confused by proximity to execve. | |
| 306 | Do need a 'the' before 'setting'. | |
| 311 | Perhaps: s/current status of tracing/current tracing status/ | |
| 318 | No 'the' needed here. | |
| 401 | Probably: s/for the process already traced/for a process already being traced/ | |
| 411 | Probably s/the non-current process/a non-current process/ | |
| 413 | And s/integer/the integer/ -- and I might actually name 'data' explicitly here to avoid confusion. | |
| 419 | Do you also need an hwpmc(4) Xref here? And, if you add DTrace, one of those as well? | |
| sys/kern/kern_procctl.c | ||
| 332 | Should we be asserting something so that we know that following p->p_pptr->p_pid is safe? E.g., the proc lock on p at the top of the function? | |
| sys/sys/procctl.h | ||
| 44 | I find it slightly confusing that PROC_TRACE_ENABLE can disable tracing. I wonder, given the naming of the operations below as PROC_TRACE_SET_FOO, whether this should be PROC_TRACE_SET. | |