Details
Details
- Reviewers
kp - Group Reviewers
network - Commits
- rGcd5ff4e841fb: pf: use hashalloc(9) for key, id, src-node and udp-endpoint hashes
Diff Detail
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Skipped - Unit
Tests Skipped - Build Status
Buildable 71746 Build 68629: arc lint + arc unit
Event Timeline
| sys/netpfil/pf/pf.c | ||
|---|---|---|
| 1435 | .head = HASH_HEAD_LIST is not needed, but I decided to be more explicit. | |
| sys/netpfil/pf/pf.c | ||
|---|---|---|
| 1448–1449 | Initialization of .type and .head to the default values. May be omitted. | |
Comment Actions
I'm seeing panics with this patch ("panic: lock "pf_keyhash" 0xfffffe00e8dffff8 already initialized").
I believe the problem is that hashalloc() allocates unzero'd memory, and which leads to incorrect assertions on the lock, if LO_INITIALIZED happens to be set in lo_flags.
Comment Actions
That's because you have it on top of older version of D55904. In the updated version MTX_NEW is passed. For me this revision passes all pf tests.