Sam Price created a merge request:
https://gitlab.rtems.org/rtems/pkg/rtems-lwip/-/merge_requests/58
Project:Branches: TheSamPrice/rtems-lwip:mdns-on-fixed-ring to
rtems/pkg/rtems-lwip:main
Author: Sam Price
## Summary
xilinx: Bound the transmit-ring wait in the MAC filter update
xemacpsif_mld6_mac_hash_update() and xemacpsif_igmp_mac_filter_update() both
open with
while(txring->HwCnt);
which never returns if the transmit-done interrupt does not retire the
outstanding descriptors. Joining a multicast group reaches here while
descriptors queued by IPv6 duplicate address detection are still outstanding,
so mdns_resp_add_netif() simply never returns.
Observed on arm/xilinx_zynq_a9_qemu, diagnosed with gdb attached to a hung
guest:
#0 xemacpsif_mld6_mac_hash_update (...) at xemacpsif.c:582
#1 xemacpsif_mld6_mac_filter_update (...)
#2 mld6_joingroup_netif (...) at mld6.c:377
#3 mdns_resp_add_netif (...) at mdns.c:2406
Bounded, under __rtems__ so the vendor code is unchanged elsewhere. Giving up
and continuing is the right failure: what follows stops the MAC, rewrites the
hash register and restarts it, so an unretired descriptor costs one packet on a
link that has only just come up -- against a function that never returns.
This is not the whole of the problem. With it applied the join completes and
the interface then stops transmitting altogether, which is the stop/reset/
restart sequence further down this same function and wants someone who knows
the GEM ring handling. This commit turns a hang into a diagnosable fault; it
does not make multicast work.
Signed-off-by: Samuel Price <[email protected]>
Assisted-by: Claude Opus 5 (1M context) <[email protected]>
## Generative AI
<!-- If you have used AI please use the "AI Contribution" template otherwise
leave this blank see our fulls statement at
https://www.rtems.org/generative-ai/-->
<!-- Default settings, if it is a dropdown it will set after submission -->
--
View it on GitLab:
https://gitlab.rtems.org/rtems/pkg/rtems-lwip/-/merge_requests/58
You're receiving this email because of your account on gitlab.rtems.org.
Unsubscribe from this thread:
https://gitlab.rtems.org/-/namespace/22/sent_notifications/5-1lgwy7idk3f3007naovs260vw-m/unsubscribe
| Manage all notifications: https://gitlab.rtems.org/-/profile/notifications |
Help: https://gitlab.rtems.org/help
_______________________________________________
bugs mailing list
[email protected]
http://lists.rtems.org/mailman/listinfo/bugs
