[Bug 292750] Jailed bhyve passthru fails with: "bhyve: PCI device at 9/0/0 is not using the ppt(4) driver"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 292750] Jailed bhyve passthru fails with: "bhyve: PCI device at 9/0/0 is not using the ppt(4) driver""
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 292750] Jailed bhyve passthru fails with: "bhyve: PCI device at 9/0/0 is not using the ppt(4) driver""
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 292750] Jailed bhyve passthru fails with: "bhyve: PCI device at 9/0/0 is not using the ppt(4) driver""
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 292750] Jailed bhyve passthru fails with: "bhyve: PCI device at 9/0/0 is not using the ppt(4) driver""
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 27 Jan 2026 01:40:56 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=292750
Bug ID: 292750
Summary: Jailed bhyve passthru fails with: "bhyve: PCI device
at 9/0/0 is not using the ppt(4) driver"
Product: Base System
Version: 15.0-STABLE
Hardware: Any
OS: Any
Status: New
Severity: Affects Only Me
Priority: ---
Component: bhyve
Assignee: virtualization@FreeBSD.org
Reporter: leper4@protonmail.com
I run a jailed bhyve with some PCI passthru devices, this setup worked fine for
quite a while on 14-STABLE. After the upgrade to 15-STABLE passing through one
of the devices failed with:
bhyve: PCI device at 9/0/0 is not using the ppt(4) driver
Device emulation initialization error: Operation not permitted
Even though the ppt driver is attached to that device. Using non-jailed bhyve
makes this work. During some of the bisecting/testing it seemed like excluding
just that device from passthru made the other devices work, however that
specific device makes the others somewhat pointless[2].
I sort of bisected[1] vmm/bhyve changes since 14-STABLE and is the culprit
https://cgit.freebsd.org/src/commit/?id=e2d6c959b29d85a83a0c827b02f3303b115daf03
as reverting it fixes this in the jailed case.
priv_check(td, PRIV_DRIVER); most likely fails in a jail, and there is no
jail.conf allow.driver or similar property that could make that work.
[1] Using git revert with all commits touching bhyve/vmm and related files.
Mainly due to limiting downtime of the machine in question.
[2] The device is an AMD APU (integrated graphics) with a VBIOS ROM.
--
You are receiving this mail because:
You are the assignee for the bug.