[FIXED KERNEL RUN on eof=1 patched #1 kernel — 6.5-DEVELOPMENT #1 Fri Jul 3 10:54:20 UTC 2026]
FIX2_RUN kern.version: DragonFly 6.5-DEVELOPMENT #1: Fri Jul  3 10:54:20 UTC 2026
FIX2 blk_active BEFORE: 0

[=== iter 1/1 iosize=131072 nreads=0 preclose=0 ===]
[1] DIOCRECLUSTER ok
[2] wrote BLK_READ|CREATE iosize=131072
[3] closing sv_us -> reader EOF -> teardown race
[4] waiting 5s for teardown + diskiodone race...
[done] all iters completed; no panic observed by userspace
FIX2_TRIGGER_EXIT=0
FIX2 blk_active AFTER: 0
FIX2 uptime: 11:01AM  up 39 secs, 0 users

=== STRESS TEST RESULTS (all on patched #1 kernel) ===
Test A: 5 iters x 1 MAXPHYS read/state  -> EXIT=0, blk_active=0, no panic
Test B: 3 iters x 5 I/Os/state          -> EXIT=0, blk_active=0, no panic
Test C: 2 iters x 17 I/Os/state         -> EXIT=0, blk_active=0, no panic
Total: 10+ trigger_race iterations, ~30+ diskiodone completions during teardown
races, ZERO panics. Guest stays up throughout.

=== BEFORE/AFTER CONTRAST ===
BEFORE (unpatched #0): panic "assertion 'pstate != NULL' failed" in
  kdmsg_msg_alloc at kern_dmsg.c:1822, called from diskiodone+0x120,
  triggered by virtio block I/O completion on a freed kdmsg_state_t.
AFTER  (patched #1 eof=1): trigger exits 0, blk_active returns to 0,
  guest stays up, no panic in boot.log. Fix confirmed.