Sure, sending v3 to add more details about the ABI.
On Mon, Jan 26, 2026 at 4:12 PM Andrew Lunn <[email protected]> wrote: > > On Mon, Jan 26, 2026 at 09:51:10PM +0100, Eric Dumazet wrote: > > On Mon, Jan 26, 2026 at 9:46 PM Jacob Keller <[email protected]> > > wrote: > > > > > > > > > > > > On 1/26/2026 9:53 AM, Andrew Lunn wrote: > > > > On Mon, Jan 26, 2026 at 12:40:15PM -0500, Brian Vazquez wrote: > > > >> On Mon, Jan 26, 2026 at 11:24 AM Andrew Lunn <[email protected]> wrote: > > > >>> > > > >>> On Mon, Jan 26, 2026 at 02:46:24PM +0000, Brian Vazquez wrote: > > > >>>> The code uses the vidx for the IRQ name but that doesn't match > > > >>>> ethtool > > > >>>> reporting or netdev naming, this makes it hard to tune the device and > > > >>>> associate queues with IRQs. Sequentially requesting irqs starting > > > >>>> from > > > >>>> '0' makes the output consistent. > > > >>>> > > > >>>> Before: > > > >>>> > > > >>>> ethtool -L eth1 tx 1 combined 3 > > > >>>> > > > >>>> grep . /proc/irq/*/*idpf*/../smp_affinity_list > > > >>>> /proc/irq/67/idpf-Mailbox-0/../smp_affinity_list:0-55,112-167 > > > >>>> /proc/irq/68/idpf-eth1-TxRx-1/../smp_affinity_list:0 > > > >>>> /proc/irq/70/idpf-eth1-TxRx-3/../smp_affinity_list:1 > > > >>>> /proc/irq/71/idpf-eth1-TxRx-4/../smp_affinity_list:2 > > > >>>> /proc/irq/72/idpf-eth1-Tx-5/../smp_affinity_list:3 > > > >>>> > > > >>>> ethtool -S eth1 | grep -v ': 0' > > > >>>> NIC statistics: > > > >>>> tx_q-0_pkts: 1002 > > > >>>> tx_q-1_pkts: 2679 > > > >>>> tx_q-2_pkts: 1113 > > > >>>> tx_q-3_pkts: 1192 <----- tx_q-3 vs idpf-eth1-Tx-5 > > > >>>> rx_q-0_pkts: 1143 > > > >>>> rx_q-1_pkts: 3172 > > > >>>> rx_q-2_pkts: 1074 > > > >>>> > > > >>>> After: > > > >>>> > > > >>>> ethtool -L eth1 tx 1 combined 3 > > > >>>> > > > >>>> grep . /proc/irq/*/*idpf*/../smp_affinity_list > > > >>>> > > > >>>> /proc/irq/67/idpf-Mailbox-0/../smp_affinity_list:0-55,112-167 > > > >>>> /proc/irq/68/idpf-eth1-TxRx-0/../smp_affinity_list:0 > > > >>>> /proc/irq/70/idpf-eth1-TxRx-1/../smp_affinity_list:1 > > > >>>> /proc/irq/71/idpf-eth1-TxRx-2/../smp_affinity_list:2 > > > >>>> /proc/irq/72/idpf-eth1-Tx-3/../smp_affinity_list:3 > > > >>>> > > > >>>> ethtool -S eth1 | grep -v ': 0' > > > >>>> NIC statistics: > > > >>>> tx_q-0_pkts: 118 > > > >>>> tx_q-1_pkts: 134 > > > >>>> tx_q-2_pkts: 228 > > > >>>> tx_q-3_pkts: 138 <--- tx_q-3 matches idpf-eth1-Tx-3 > > > >>>> rx_q-0_pkts: 111 > > > >>>> rx_q-1_pkts: 366 > > > >>>> rx_q-2_pkts: 120 > > > >>> > > > >>> Are there any ABI issues here? > > > >> > > > >> The patch doesn't change the format, it just fixes the numbering in > > > >> the name to make it consistent with other reporting tools. It > > > >> shouldn't break any library. > > > > > > > > But is the numbering part of the ABI? > > > > > > > > Making a comment about ABI in the commit message makes it clear it is > > > > something you have considered, and you have decided it is not an > > > > issue. If there is no such comment, reviewers probably should ask. > > > > > > > > Andrew > > > > > > I don't see how an application can depend on the name if it can't > > > correlate it to anything meaningful. The change fixes the ID values used > > > so that they *do* correlate. If an application was previously assuming > > > it correlated to the queue ID, it would incorrect associate the IRQ with > > > the wrong queue. > > > > Agree, this mismatch caused me some surprises, we can not keep a broken > > ABI (which has been broken by accident) > > So everybody is agreed, great. I just think it would of been good to > mention ABI in the commit message, to show due diligence has been > done. > > Andrew
