# DF-1599 — PoC Verification Verdict

**Category:** radeon (module, HW-gated)
**Source:** `sys/dev/drm/radeon/radeon_vm.c:453-466`
**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

radeon_vm_bo_set_addr declares last_pfn as 'unsigned' (32-bit) at line 453, but eoffset is uint64_t. At line 465 'last_pfn = eoffset / RADEON_GPU_PAGE_SIZE' truncates to low 32 bits; the bounds check at 466 compares the TRUNCATED value against max_pfn. A huge offset (e.g. 2^48) truncates to a small value that passes the check, allowing VA above the limit.

**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

Change 'unsigned last_pfn' to 'uint64_t last_pfn' so the division and comparison are 64-bit.

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.