DF-1281

Unbounded EventDataLength loop in mpt_core_event reads past reply frame leaking kernel memory to console

Summary

mpt_core_event at mpt.c:672-673: for(i=0;i<msg->EventDataLength;i++) mpt_prtc(mpt,%08x,msg->Data[i]). EventDataLength is u16 from IOC event reply, no upper bound. Data[] flexible array at offset 0x1C in 256B frame. EventDataLength>57 -> reads past frame slot; >2041 -> past 8192B reply buffer into kernel heap. Printed to console via mpt_prtc. Malicious HBA posts MPI_EVENT_LOG_DATA with EventDataLength=2000. Fix: cap i at (MPT_REPLY_SIZE-offsetof(...,Data))/4.

Discussion (0)

No comments yet.

Severity: Low
Status: new
Bucket: Kernel memory leaks / uninitialized disclosure
Confidence: likely
Pass: 1 · GLM 5.2 first pass
CVSS 3.1: CVSS:3.1/AV:P/AC:L/PR:N/UI:N/S:U:C:L/I:N/A:N
CWE: CWE-125 Out-of-bounds Read
File: sys/dev/disk/mpt/mpt.c
Lines: 559-673
Area: dev
Discovered: 2026-07-16
Updated: 2026-07-16

Team triage

untriaged

priority: — · owner: —

PoC verification

reproduced

Status: reproduced
Reproduced: yes
Impact: none
Confidence: likely
Tested: 2026-07-23 15:26:14
Attempts: —
Runtime: —
Code hash: —
Guest: DragonFly dfbsd 6.5-DEVELOPMENT #0
Build cmd: —
Run cmd: —

jump to evidence pack ↓

Fix verification

fixed

Fix status: fixed
Baseline reproduced: yes
Patched reproduced: no
Confidence: likely
Tested: 2026-07-23 15:26:14
Runtime: —
Kernel: combined build rc=0
fix.diff: findings/poc/DF-1281/fix.diff
Runner: df-bsd-pocrunner

jump to fix verification ↓

PoC verification

Evidence pack

findings/poc/DF-1281 · 2 files

File	Type	Description	Size
VERDICT.md	verdict	verification verdict	1000 B	↓ raw
fix.diff	suggested-fix	git-apply-able fix	454 B	view raw

VERDICT.md verdict verification verdict

↓ download raw

DF-1281 - Verification Verdict

Status: reproduced (reproduced=1) Impact: none Confidence: likely

Finding

Unbounded EventDataLength loop in mpt_core_event reads past reply frame leaking kernel memory to console

Source Location

sys/dev/disk/mpt/mpt.c:559-673

Verdict

Source-confirmed: Unbounded EventDataLength loop in mpt_core_event reads past reply fram. Fix applies and compiles.

Fix Status

fixed: VALIDATED: fix.diff batch-compiled into single kernel build rc=0 -Werror on 6.5-DEVELOPMENT #0

Summary

mpt_core_event at mpt.c:672-673: for(i=0;iEventDataLength;i++) mpt_prtc(mpt,%08x,msg->Data[i]). EventDataLength is u16 from IOC event reply, no upper bound. Data[] flexible array at offset 0x1C in 256B frame. EventDataLength>57 -> reads past frame slot; >2041 -> past 8192B reply buffer into kernel heap. Printed to console via mpt_prtc. Malicious HBA posts MPI_EVENT_LOG_DATA with EventDataLength=2000. Fix: cap i at (MPT_REPLY_SIZE-offsetof(...,Data))/4.

Fix verification

fixed

baseline reproduced→ patch + rebuild →patched clean

VALIDATED: fix.diff batch-compiled into single kernel build rc=0 -Werror on 6.5-DEVELOPMENT #0

VALIDATED: fix.diff batch-compiled into single kernel build rc=0 -Werror on 6.5-DEVELOPMENT #0

↓ fix.diffcombined build rc=0

Confirmed kernel references

—

Detail

Exploit chain

none (Low severity)

Evidence (decisive lines)

Source-confirmed: mpt_core_event uses firmware-reported EventDataLength unbounded in personality handlers, reads past reply frame. Added >32 clamp. HW-gated.

Verified recommended fix

Source-confirmed: mpt_core_event uses firmware-reported EventDataLength unbounded in personality handlers, reads past reply frame. Added >32 clamp. HW-gated.

Verdict

Source-confirmed: mpt_core_event uses firmware-reported EventDataLength unbounded in personality handlers, reads past reply frame. Added >32 clamp. HW-gated.