> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) > to consider: > - [Medium] User-configured RSS indirection table is silently discarded and > reset to defaults during per-queue reconfigurations. > -- > > --- Patch [1]: [PATCH net-next v2 01/13] net: mana: add queue-set allocation > and teardown helpers --- commit > fcc6a74602d4e66dc287809434c5e9b44c3be829 > Author: Long Li <[email protected]> > Subject: net: mana: add queue-set allocation and teardown helpers > > The ethtool reconfiguration paths (channel count, ring size, private flags), > mana_change_mtu() and mana_xdp_set() rebuild the queues with > mana_detach() then mana_attach(). This adds the data model and helpers > for pre-allocate and swap to independently build and tear down a queue set. > > > diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c > > b/drivers/net/ethernet/microsoft/mana/mana_en.c > > --- a/drivers/net/ethernet/microsoft/mana/mana_en.c > > +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c > [ ... ] > > +int mana_alloc_qset(struct mana_port_context *scratch, unsigned int > num_queues, > > + unsigned int rx_queue_size, unsigned int tx_queue_size, > > + u32 priv_flags, struct mana_qset *out) { > > + struct net_device *ndev = scratch->ndev; > > + int err; > [ ... ] > > + err = mana_rss_table_alloc(scratch); > > + if (err) > > + goto cleanup_rxq_array; > > + > > + err = mana_create_eq(scratch); > > + if (err) > > + goto cleanup_rss; > > + > > + err = mana_create_txq(scratch, ndev); > > + if (err) > > + goto cleanup_eq; > > + > > + err = mana_add_rx_queues(scratch, ndev); > > + if (err) > > + goto cleanup_rxq; > > + > > + mana_rss_table_init(scratch); > > [Severity: Medium] > Will this unconditionally reset the RSS indirection table to defaults during > reconfigurations? > > Since mana_rss_table_alloc() creates a new table and mana_rss_table_init() > initializes it to default round-robin values, it seems like a user's existing > custom RSS steering configuration might be silently overwritten when the new > queue set is published. > > Could this cause a regression where the custom configuration is lost? Should > it be preserved by copying the existing table from the active port context > instead?
patch 11/13 fixed this. Long
