On 6/29/26 13:39, Aditya Gupta wrote:
On 26/06/29 01:20PM, Cédric Le Goater wrote:
On 6/29/26 13:04, Aditya Gupta wrote:
In the past there have been hard to recreate issues where XIVE changes cause qemu crashes due to multi-socket interrupts such as in [1].
Add a functional test explicitly to test whether remote interrupts work.
The test can also work as additional boot test for multi-socket boot, initrd boot test, as well as a check for e1000e to be working in powernv,
you should consider moving to a more modern nic model: igb, which has sr-iov support too.
yes, can do. with igb i see multiple irqs, so test will have to take sum of all rx irqs, and compare:
2026-06-29 17:00:46,288: cat /proc/interrupts | grep eth0 2026-06-29 17:00:46,391: 135: 0 0 0 PNV-PCI-MSI 2416443392 Edge eth0 2026-06-29 17:00:46,393: 136: 0 6 0 PNV-PCI-MSI 2416443393 Edge eth0-rx-0 2026-06-29 17:00:46,394: 137: 0 14 0 PNV-PCI-MSI 2416443394 Edge eth0-rx-1 2026-06-29 17:00:46,395: 138: 0 4 0 PNV-PCI-MSI 2416443395 Edge eth0-rx-2 2026-06-29 17:00:46,397: 139: 0 4 0 PNV-PCI-MSI 2416443396 Edge eth0-rx-3 2026-06-29 17:00:46,398: 140: 0 16 0 PNV-PCI-MSI 2416443397 Edge eth0-tx-0 2026-06-29 17:00:46,399: 141: 0 6 0 PNV-PCI-MSI 2416443398 Edge eth0-tx-1 2026-06-29 17:00:46,399: 142: 0 6 0 PNV-PCI-MSI 2416443399 Edge eth0-tx-2 2026-06-29 17:00:46,400: 143: 0 4 0 PNV-PCI-MSI 2416443400 Edge eth0-tx-3
I feel e1000e suffices for this test, should i stay with e1000e or use igb here ?
or use a single cpu per socket config. Which seems better for the test. Thanks, C.