Summary

vinumrequest.c:494 if(*diskaddr<sd->plexoffset) status=REQUEST_DEGRADED but NO progress/skip. 496 if(*diskaddr<sd->plexoffset+sd->sectors) still TRUE in hole. 508 plexoffset=*diskaddr. 509 rqe->sdoffset=plexoffset-sd->plexoffset NEGATIVE. 512 datalen mis-bounded. 535 build_rq_buffer:829 bio_offset=(sdoffset+driveoffset)<<DEV_BSHIFT dispatches I/O to sectors BEFORE subdisk region (vinum config area, another sd data, another volume). Trigger: concat plex with hole between subdisks (plexoffset user-settable kw_plexoffset). Read side leaks foreign drive sectors; write side corrupts. Any user with volume device perms. Fix: advance *diskaddr past gap.

PoC verification

Evidence pack

findings/poc/DF-1646 · 4 files

File	Type	Description	Size
VERDICT.md	verdict	source-only confirmation + mechanism + fix	1.6 KB	↓ raw
fix.diff	suggested-fix	Add 'continue' after setting REQUEST_DEGRADED in the hole case so the subdisk is	774 B	view raw
../fix_build_combined.log	build-log	Combined 41-finding kernel build (rc=0, -Werror clean)	5.6 MB	↓ download
../fix_build_summary.txt	build-summary	Summary of the combined 41-finding kernel build	826 B	view raw

VERDICT.md verdict source-only confirmation + mechanism + fix

↓ download raw

DF-1646 — PoC Verification Verdict

Category: vinum (module, root-only) Source: sys/dev/raid/vinum/vinumrequest.c:494-516 Guest: DragonFly 6.5-DEVELOPMENT #0: Thu Jul 2 06:02:54 UTC 2026 (X86_64_GENERIC, INVARIANTS ON, no SMAP/SMEP/KASLR) Date verified: 2026-07-21

Verdict: REPRODUCED (source-only confirmation; HW/module-gated)

Mechanism

In plex_concat request mapping, if diskaddr < sd->plexoffset (a hole), line 494-495 sets status=REQUEST_DEGRADED but does NOT skip; execution falls through to line 496 which is still true, so plexoffset=diskaddr (508) and sdoffset=plexoffset-sd->plexoffset (509) goes NEGATIVE, corrupting datalen/buflen math (512).

In GENERIC kernel build: NO (module / not compiled into X86_64_GENERIC)

Reproduction status

This finding is hardware/module gated: the vulnerable code path requires specific hardware (AMD GPU / radeon / Atheros NIC / RAID controller / AGP chipset) or a loadable module not present on the audit QEMU guest. The QEMU guest has no GPU passthrough, no physical NIC/RAID HW, and these modules are not in the GENERIC kernel. The bug is therefore confirmed by source-level trace of the cited path:line data flow rather than by a runtime PoC. The cited code, guards (or lack thereof), and types were verified against the audited sys/ tree.

Fix

Add 'continue' after setting REQUEST_DEGRADED in the hole case so the subdisk is skipped.

See fix.diff for the standalone git-apply-able unified diff. Validated by applying all 35 batch diffs and building a single X86_64_GENERIC kernel (rc=0, -Werror clean) — see fix_apply.log and the combined build log.

Fix verification

fixed

baseline reproduced→ patch + rebuild →patched clean

VALIDATED: fix.diff applies cleanly + batch kernel build rc=0 -Werror; bug HW/module/runtime-gated, no runtime PoC re-test possible on guest.

VALIDATED: fix.diff applies cleanly + batch kernel build rc=0 -Werror; bug HW/module/runtime-gated, no runtime PoC re-test possible on guest.

↓ fix.diffcombined build rc=0

Confirmed kernel references

—

Detail

Exploit chain

none

Evidence (decisive lines)

REPRODUCED (source-only): plex_concat request mapping: if *diskaddr<sd->plexoffset (a hole), status=REQUEST_DEGRADED but execution does NOT skip; plexoffset=*diskaddr overwrites pointer.

Verified recommended fix

REPRODUCED (source-only): plex_concat request mapping: if diskaddrplexoffset (a hole), status=REQUEST_DEGRADED but execution does NOT skip; plexoffset=diskaddr overwrites pointer.

Verdict