β¬’ DragonFlyBSD Kernel Audit
← triage Β· dashboard
DF-0961

sys_sbrk does not enforce RLIMIT_VMEM (address-space limit), unlike sys_obreak

Summary

sys_sbrk enforces only RLIMIT_DATA at vm_unix.c:116-121 before vm_map_find, never checks RLIMIT_VMEM. Sibling sys_obreak correctly checks RLIMIT_VMEM at :218-222 (vm->vm_map.size + diff > rlim_cur -> ENOMEM). RLIMIT_VMEM is real kernel-enforced limit (checked at exec imgact_elf.c:793-794 and in obreak). Process with RLIMIT_DATA > RLIMIT_VMEM (uncommon but valid containment config) can exceed VMEM cap via sbrk(). No priv esc/mem corruption (RLIMIT_DATA still caps to maxdsiz=32GB, vm_map_insert rejects overlaps). Fix: mirror the RLIMIT_VMEM check before vm_map_find.

Discussion (0)

No comments yet.

PoC verification

Evidence pack

findings/poc/DF-0961 Β· 2 files
FileTypeDescriptionSize
fix.diff suggested-fix git-apply-able fix for the cited path 320 B view raw
VERDICT.md verdict source-confirmation narrative 943 B ↓ raw
VERDICT.md verdict source-confirmation narrative
↓ download raw

DF-0961 source-confirmation

Verdict: REPRODUCED (source-confirmed)
Impact: none Confidence: likely

Kernel ref: sys/vm/vm_unix.c:116

Mechanism

sys_sbrk checks RLIMIT_DATA but never RLIMIT_VMEM

Confirmation method

Source-trace confirmed the cited code path matches the finding (exact line/condition verified against sys/). Runtime PoC not exercised for this source-only Low-severity item; confirmation is by code inspection.

See fix.diff in this folder (git-apply-able unified diff).

Phase 8 (combined build)

All 64 fixes were batched into one combined patch (../_batch/combined_64.patch) and applied to in-guest /usr/src. A single make -j6 nativekernel KERNCONF=X86_64_GENERIC completed rc=0 with 0 errors under -Werror (../_batch/fix_build.log, 35374 lines). The GENERIC kernel + all modules (drm, firewire, usb, netgraph, smbfs, fuse, crypto, vm, pmap) compiled clean.

Fix verification

not_testable
baseline reproduced→ patch + rebuild →patched clean

not_testable (source-only): no runtime PoC exercised; fix.diff applies cleanly and the combined single nativekernel build of all 64 fixes completed rc=0 with 0 errors under -Werror (findings/poc/_batch/fix_build.log). Bug source-confirmed in baseline tree. Compile-validation of the fix is the requested Phase-8 deliverable for this batch.

combined build: '=== NK_DONE rc=0 ===' and '>>> Kernel build for X86_64_GENERIC completed on Thu Jul 23 09:17:55 UTC 2026' (0 'error:' lines in 35374-line log).
↓ fix.diffDragonFly 6.5-DEVELOPMENT single-fix combined kernel (nativekernel rc=0 -Werror, 2026-07-23 09:17:55 UTC)

Confirmed kernel references

Detail

Exploit chain

none (non-corruption / source-only confirmation; no memory-corruption escalation chain developed for this Low-severity item)

Evidence (decisive lines)

source-only: sys_sbrk checks RLIMIT_DATA but never RLIMIT_VMEM @ sys/vm/vm_unix.c:116. Combined fix build: '=== NK_DONE rc=0 ===' / 'Kernel build for X86_64_GENERIC completed', 0 errors under -Werror (findings/poc/_batch/fix_build.log).

PoC changes

authored git-apply-able fix.diff targeting the cited line; source-trace verified the vulnerable path. Evidence pack (VERDICT.md, manifest.json, env.txt, fix.diff) in findings/poc/DF-0961/.

Verified recommended fix

mirror the RLIMIT_VMEM check before vm_map_find. matches finding proposal. Full diff in findings/poc/DF-0961/fix.diff.

Verdict

REPRODUCED (source-confirmed). Source-trace confirmed sys_sbrk checks RLIMIT_DATA but never RLIMIT_VMEM at sys/vm/vm_unix.c:116 against the audited sys/ tree; exact line/condition verified.