[PATCH net v3 0/2] bonding: fix TLB load-tracking overflow on high-speed NICs
Hangbin Liu <[email protected]>
| Newsgroups | gmane.linux.network,gmane.linux.kernel |
|---|---|
| Message-ID | <[email protected]> |
The bonding TLB (Transmit Load Balancing) mode tracks per-slave and per-client transmit byte counts in u32 fields. At sustained throughput above ~3.2 Gbit/s over the 10-second rebalance interval these counters wrap, causing compute_gap() to produce incorrect gap values and mis-select transmit slaves. Such speeds are routine on modern NICs under heavy traffic. This two-patch series fixes the overflow by widening the relevant fields to u64. Patch 1 converts the unbalanced_load counter to per-cpu state as a preparatory step. The counter sits in the transmit hot path, so converting it to per-cpu before widening avoids introducing cross-CPU synchronization overhead for a u64. Patch 2 widens tx_bytes, load_history, load, and the per-cpu unbalanced_load tx_bytes from u32 to u64. It adds u64_stats_sync protection for the per-cpu counter to prevent tearing on 32-bit architectures, and reworks compute_gap() to use u64 arithmetic with READ_ONCE() on slave->speed. Note: In patch 2, I return 0 directly when the speed is unknown, because I believe the comparison becomes meaningless if the NIC speed cannot be determined. For example: a 1 G NIC and a 10 G NIC are bonded together, yet both are marked as unknown. If you would prefer to retain the existing logic, we could instead set the speed to `ULLONG_MAX` when speed is unknown. Signed-off-by: Hangbin Liu <[email protected]> --- Changes in v3: - Add a preparatory patch to convert unbalanced_load to per-cpu first - widens tlb counters to u64 and add helpers to prevent tearing on 32-bit - Link to v2: https://lore.kernel.org/r/[email protected] Changes in v2: - update comment description, including AI-detected info. - fix tx_bytes/load type detected by sashiko - cast SPEED_UNKNOWN to 0 before shift, detected by sashiko - Link to v1: https://lore.kernel.org/r/[email protected] --- Hangbin Liu (2): bonding: convert unbalanced_load to per-cpu state bonding: fix u32 overflow in compute_gap() drivers/net/bonding/bond_alb.c | 69 +++++++++++++++++++++++++++++++---------- drivers/net/bonding/bond_main.c | 9 ++++++ include/net/bond_alb.h | 13 +++++--- 3 files changed, 71 insertions(+), 20 deletions(-) --- base-commit: 21040c7f931502070dcc66bb0f1aeed07dec032b change-id: 20260806-bond_overflow-ac6a6a78d6a0 Best regards, -- Hangbin Liu <[email protected]>