Repository navigation
anxzz9000: program the GEM multicast hash - #202
Merged
Merged
Conversation
Upstream firmware (BlitterStudio PR #111) starts the GEM with an empty multicast hash and takes bucket commands at ETH_CONFIG (0x8a). The driver never wrote them: it relied on the fork firmware accepting every group. On upstream firmware no multicast frame arrived, so IPv6 neighbour discovery for the card's addresses failed (multicast in 0 on the A3000). When the firmware reports the capability, rebuild the hash at init and on every filter change: one bucket per joined group, or all 64 when promiscuous or all-multicast. The software filter stays, as the buckets collide. The hash is the GEM's, as in ZZ9000Net.device's net/mcast.c. Co-Authored-By: Claude Opus 5.5 <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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Upstream ZZ9000 firmware (BlitterStudio zz9000-firmware #111) starts the GEM with an empty multicast hash and takes bucket commands at ETH_CONFIG 0x8a. anxzz9000.device never wrote them, relying on the old fork firmware accepting every group. On the A3000 (upstream/all firmware) eth0 counted multicast in 0 and IPv6 neighbour discovery for the ZZ9000 addresses failed.
When the firmware reports the capability, the hash is rebuilt at init and on each filter change: one bucket per joined group, all 64 when promiscuous or all-multicast. The software filter stays. Hash matches ZZ9000Net.device net/mcast.c; checked against the GEM bit-fold over 10000 random addresses.
On the A3000 (2026-10-08 02:40Z, this build sha256 fdbc7a98): NDP REACHABLE for the ZZ9000 global and link-local addresses, HTTP 200 over both, multicast in 11 (was 0). anxzz9000.device is not in the release archive, so no CHANGELOG entry.
🤖 Generated with Claude Code