Alan pointed out that the object resides in the global object list, so
formally we should lock it to synchronize with a scan of the list.
Details
Details
- Reviewers
alc kib - Commits
- rS354402: Convert ksyms(4) to use an OBJT_PHYS object.
Diff Detail
Diff Detail
- Repository
- rS FreeBSD src repository - subversion
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
Comment Actions
I realized after submitting this that really we should be using OBJT_PHYS here. The object's contents are not pageable.
sys/dev/ksyms/ksyms.c | ||
---|---|---|
44 ↗ | (On Diff #63911) | Is this still needed? |
Comment Actions
Yes, I believe so. However, in recent weeks, I recall seeing at least two attempts to reintroduce unlocked manipulations, counting this as one of them. I think that a vm_object_allocate_flags() would eliminate the temptation to do so.