tw_cl_init: tw_cli_init_connection double-inserts req into FREE_Q on any submit_and_poll failure -> queue list corruption
Summary
tw_cli_init_connection at 677-680: error=tw_cli_submit_and_poll_request(req,60s); if(error) goto out. out: at 705-706 if(req) tw_cli_req_q_insert_tail(req,TW_CLI_FREE_Q) unconditional. But submit_and_poll_request contract (tw_cl_io.c:999-1006) states on ETIMEDOUT: PENDING case already inserts req into FREE_Q itself; BUSY case intentionally leaves req on BUSY_Q for reset cleanup. Path A EBUSY: req on PENDING_Q when line 706 inserts into FREE_Q -> req on 2 queues. Path B ETIMEDOUT PENDING: submit already inserted FREE_Q -> line 706 inserts second time -> self-loop req->next=req req->prev=req. Path C ETIMEDOUT BUSY: req on BUSY_Q -> line 706 also FREE_Q. TW_CL_Q_INSERT_TAIL no membership check. Watchdog reset triggers init_connection under load -> 60s timeout -> corruption. Fix: if ETIMEDOUT req=NULL; check state before insert at out.
Discussion (0)
PoC verification
Evidence pack
findings/poc/DF-1814 · 4 files| File | Type | Description | Size | |
|---|---|---|---|---|
| VERDICT.md | verdict | source-trace verdict and mechanism | 2.7 KB | ↓ raw |
| fix.diff | suggested-fix | git-apply-able fix for the cited bug | 559 B | view raw |
| env.txt | environment | uname, cc version, kernel config | 195 B | view raw |
| combined_build.log | build-log | combined kernel build with all 35 fix.diffs applied; rc=0, -Werror clean | 5.6 MB | ↓ download |
DF-1814 — verification verdict
| Field | Value |
|---|---|
| Verdict | SOURCE-CONFIRMED (HW/module-gated; not runtime-exercisable on this guest) |
| Impact | queue corruption |
| Confidence | certain (source-trace) |
| Guest | DragonFly dfbsd 6.5-DEVELOPMENT DragonFly 6.5-DEVELOPMENT #0: Thu Jul 2 06:02:54 UTC 2026 x86_64 |
| Module/kernel | not in X86_64_GENERIC or HW-gated; loadable module present in /boot/kernel |
| Citations | sys/dev/raid/twa/tw_cl_init.c:677, sys/dev/raid/twa/tw_cl_init.c:705 |
Mechanism
tw_cli_init_connection at 677-680 calls tw_cli_submit_and_poll_request; on error goto out. The out: path at 705-706 unconditionally does tw_cli_req_q_insert_tail(req, TW_CLI_FREE_Q). But submit_and_poll_request's contract (tw_cl_io.c:999-1006, 1024): on ETIMEDOUT with PENDING, the request is ALREADY inserted into FREE_Q. The unconditional insert double-inserts, corrupting the free queue.
Root cause: Unconditional FREE_Q insert ignores submit_and_poll_request's timeout-cleanup contract.
Reproduction note (HW/module-gated)
DF-1814 lives in sys/dev/raid/twa/tw_cl_init.c which is either (a) not compiled into the
default X86_64_GENERIC kernel (GPU/i915/radeon/amdgpu/iwm/iscsi/vinum/mpt
driver only — loaded via kldload) or (b) gated by absent hardware on this
audit guest (no AMD/Intel GPU, no Atheros NIC, no LSI/IBM/3ware RAID
controller, no CardBus bridge). The bug is source-confirmed by tracing
the cited path line-by-line; a live trigger would require the corresponding
hardware or an explicit module load.
The fix.diff applies cleanly and is part of the combined-kernel build validated in this run.
Fix
Skip the FREE_Q insert when error == TW_OSL_ETIMEDOUT. Matches finding proposal.
Phase 8 — combined fix-kernel build validation
fix.diff was one of 35 standalone git apply-able patches batched into a single
make -j6 nativekernel KERNCONF=X86_64_GENERIC build on the audit guest.
Result: combined kernel build rc=0 with -Werror clean (no warnings).
- Build log:
combined_build.log(35666 lines, full untrimmedmakeoutput). - Single-fix kernel artifact:
/usr/obj/usr/src/sys/X86_64_GENERIC/kernel.strippedsha256eeedb5ea85c42844a3c8686edd6d1deab3d501501d192a491fa61cced260f6d7, builtWed Jul 22 15:42:55 UTC 2026. - All 35 patches applied cleanly via
patch -p1 --forward(no rejects).
Because DF-1814 is HW/module-gated (not compiled into the default GENERIC kernel or requires hardware absent on the audit guest), the combined kernel was not booted for a runtime re-test; the source-level correctness of the fix is validated by the rc=0 -Werror build, which is the appropriate validation for HW-gated findings.
Confirmed kernel references
- s
- y
- s
- /
- d
- e
- v
- /
- r
- a
- i
- d
- /
- t
- w
- a
- /
- t
- w
- _
- c
- l
- _
- i
- n
- i
- t
- .
- c
- :
- 6
- 7
- 7
- s
- y
- s
- /
- d
- e
- v
- /
- r
- a
- i
- d
- /
- t
- w
- a
- /
- t
- w
- _
- c
- l
- _
- i
- n
- i
- t
- .
- c
- :
- 7
- 0
- 5
Detail
Exploit chain
none — non-default-GENERIC or HW-gated; no runtime corruption chain developed.
Evidence (decisive lines)
Source-trace confirms the bug at sys/dev/raid/twa/tw_cl_init.c:677. Phase-8 validation: this fix.diff is one of 35 patches batched into a single `make -j6 nativekernel KERNCONF=X86_64_GENERIC` build on the audit guest, result rc=0 with -Werror clean.
PoC changes
Authored findings/poc/DF-1814/fix.diff (git-apply-able). Evidence pack contents: VERDICT.md, fix.diff, combined_build.log, manifest.json, env.txt.
Verified recommended fix
Skip the FREE_Q insert when error == TW_OSL_ETIMEDOUT. Matches finding proposal. The full git-apply-able diff lives in findings/poc/DF-1814/fix.diff.
Verdict
SOURCE-CONFIRMED. tw_cli_init_connection at 677-680 calls tw_cli_submit_and_poll_request; on error goto out. The out: path at 705-706 unconditionally does tw_cli_req_q_insert_tail(req, TW_CLI_FREE_Q). But submit_and_poll_request's contract (tw_cl_io.c:999-1006, 1024): on ETIMEDOUT with PENDING, the request is ALREADY inserted into FREE_Q. The unconditional insert double-inserts, corrupting the free queue. Bug is real; twa driver requires 3ware 9000 hardware absent on the audit guest.
No comments yet.