fix: validate atomic target addr against heap range in proxy DrainCq - #23
Merged
itej89 merged 1 commit intoAug 28, 2026
Merged
Conversation
Native IBV_WR_ATOMIC_FETCH_AND_ADD validates the target address via the MR rkey. The SEND_WITH_IMM emulation drops that — the CPU does an atomic wherever the payload points, so wire corruption becomes an arbitrary write. Pass heapBaseAddr/heapEndAddr from GpuStates into ProxyThread::Init and reject payload.addr outside [heapBase, heapEnd) in DrainCq. Addresses PR ROCm#558 round-2 review comment R2-1. Co-Authored-By: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Addresses PR ROCm#558 round-2 review comment R2-1.
Native
IBV_WR_ATOMIC_FETCH_AND_ADDvalidates the target address via the MR rkey. TheSEND_WITH_IMMemulation drops that — the CPU does an atomic wherever the payload points, so wire corruption becomes an arbitrary write.Fix: pass
heapBaseAddr/heapEndAddrfromGpuStatesintoProxyThread::Initand rejectpayload.addroutside[heapBase, heapEnd)inDrainCq. When heap range is not set (both zero), the check is skipped for backward compatibility.Test plan
🤖 Generated with Claude Code