Re: kern/59165: vio9p(4): expose tag through sysctl or device properties

["Greg A. Woods via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR kern/59165; it has been noted by GNATS.

From: "Greg A. Woods" <woods%planix.ca@localhost>
To: NetBSD-current Users Discussion List <current-users%netbsd.org@localhost>
Cc: NetBSD Kernel Technical Discussion List <tech-kern%NetBSD.org@localhost>,
    NetBSD GNATS <gnats-bugs%NetBSD.org@localhost>
Subject: Re: kern/59165: vio9p(4): expose tag through sysctl or device properties
Date: Tue, 25 Nov 2025 17:39:19 -0800

 --pgp-sign-Multipart_Tue_Nov_25_17:39:11_2025-1
 Content-Type: text/plain; charset=US-ASCII
 
 I was looking into when and why vio9p(4) started working (on amd64 I
 have it working in 10.99.12, as of early April) and I ran into mention
 of a sysctl to see the tag names, and then I found PR#59165.
 
 There's no mention of this sysctl node in the viop(4) manual page
 though, nor in sysctl(7) (though I wouldn't expect it in the latter).
 
 I see the code is there in sys/dev/pci/vio9p.c, but it doesn't seem to
 work for me:
 
 
 $ uname -a
 NetBSD nbt2.local 10.99.12 NetBSD 10.99.12 (GENERIC) #1: Tue Apr  8 14:29:37 PDT 2025  woods@very.local:/Users/woods/build/woods/very.local/trunk-x86_64-amd64-obj/Volumes/work/woods/g-NetBSD-src/sys/arch/amd64/compile/GENERIC amd64
 $ /sbin/modstat | fgrep vio9p
 vio9p                      driver   builtin  -        0       - virtio
 $ fgrep vio9p /var/run/dmesg.boot
 vio9p0 at virtio3: features: 0x10000001<INDIRECT_DESC,MOUNT_TAG>
 virtio3: allocated 24576 byte for virtqueue 0 for vio9p, size 128
 vio9p0: tagged as share
 $ /sbin/sysctl hw.vio9p
 sysctl: second level name 'vio9p' in 'hw.vio9p' is invalid
 $ /sbin/sysctl -a | fgrep vio9p
 kern.drivers = [[...]], [356 -1 vio9p], [[...]]
 
 I think the code looks right, and given I see the output of the device
 printf that's executed right above that code, well....
 
 For the record I'm running this under QEMU, with UTM, on macOS(i386),
 and accessing the shared directory via 9p is working very well.
 
 Is anyone else who might be using vio9p(4) seeing this sysctl node?
 
 --
 					Greg A. Woods <gwoods%acm.org@localhost>
 
 Kelowna, BC     +1 250 762-7675           RoboHack <woods%robohack.ca@localhost>
 Planix, Inc. <woods%planix.com@localhost>     Avoncote Farms <woods%avoncote.ca@localhost>
 
 --pgp-sign-Multipart_Tue_Nov_25_17:39:11_2025-1
 Content-Type: application/pgp-signature
 Content-Transfer-Encoding: 7bit
 Content-Description: OpenPGP Digital Signature
 
 -----BEGIN PGP SIGNATURE-----
 
 iF0EABECAB0WIQRuK6dmwVAucmRxuh9mfXG3eL/0fwUCaSZaQQAKCRBmfXG3eL/0
 fwiBAKCA9CpvBt2i9YggIkscw2/G03UuOgCePiWy8kG71pZrOqGMljfSfsy4h+A=
 =oA6C
 -----END PGP SIGNATURE-----
 
 --pgp-sign-Multipart_Tue_Nov_25_17:39:11_2025-1--