On Thu, 12 Feb 2026 at 11:55, Eli Britstein <[email protected]> wrote:
> 2026-02-12T10:45:01.704Z|00013|dpdk|ERR|EAL: failed to parse device
> "auxiliary:mlx5_core.sf.0"
> 2026-02-12T10:45:01.704Z|00014|dpdk|ERR|EAL: Unable to parse device
> 'auxiliary:mlx5_core.sf.0'
Mm, as I read the auxiliary bus .parse() code, would a simple
"auxiliary:" be enough?
Otherwise, I looked at the other buses .parse()/.scan(), and ended up with:
diff --git a/lib/dpdk.c b/lib/dpdk.c
index d27b95cd9a..794ffbe599 100644
--- a/lib/dpdk.c
+++ b/lib/dpdk.c
@@ -430,6 +430,34 @@ dpdk_init__(const struct smap *ovs_other_config)
svec_add_nocopy(&args, xasprintf("0@%d", cpu));
}
+ if (!args_contains(&args, "-a") && !args_contains(&args, "-b")
+ && !smap_get_bool(ovs_other_config, "dpdk-probe-at-init", false)) {
+#ifdef RTE_BUS_AUXILIARY
+ svec_add(&args, "-a");
+ svec_add(&args, "auxiliary:");
+#endif
+#ifdef RTE_BUS_CDX
+ svec_add(&args, "-a");
+ svec_add(&args, "cdx:cdx-");
+#endif
+#ifdef RTE_BUS_FSLMC
+ svec_add(&args, "-a");
+ svec_add(&args, "fslmc:dpni.65535");
+#endif
+#ifdef RTE_BUS_PCI
+ svec_add(&args, "-a");
+ svec_add(&args, "pci:0000:00:00.0");
+#endif
+#ifdef RTE_BUS_UACCE
+ svec_add(&args, "-a");
+ svec_add(&args, "uacce:");
+#endif
+#ifdef RTE_BUS_VMBUS
+ svec_add(&args, "-a");
+ svec_add(&args, "vmbus:00000000-0000-0000-0000-000000000000");
+#endif
+ }
+
svec_terminate(&args);
optind = 1;
It seems to cover all DPDK builtin buses (iiuc, other buses .scan()
don't populate devices).
I drafted a DPDK change for disabling the whole probing at init via a
single --no-device-probe knob (for now, it would be conflicting with
any -a/-b options).
I'll probably submit it for v26.07.
--
David Marchand
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
