Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 07 Jan 2020 16:51:12 +0000
From:      bugzilla-noreply@freebsd.org
To:        virtualization@FreeBSD.org
Subject:   [Bug 241774] FreeBSD 11.3 & 12.0 has broken SCSI & Networking on KVM/QEMU Q35 with OVMF
Message-ID:  <bug-241774-27103-UdOs41QPpz@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-241774-27103@https.bugs.freebsd.org/bugzilla/>
References:  <bug-241774-27103@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D241774

Tommy P <tommyhp2@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tommyhp2@gmail.com

--- Comment #11 from Tommy P <tommyhp2@gmail.com> ---
Hi John,

I decided to post here instead of my bug report
(https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D236922) since it adds =
OVMF
testing.

I've just created a OVMF + Q35 VM for 12.0 fresh install.  Intel NIC driver
works fine for me with the original default install kernel and custom updat=
ed
kernel.  Note the 'Hypervisor default' SCSI (sym0) loaded below:

root@fbsd12-ovmf:~ # uname -a
FreeBSD fbsd12-ovmf 12.0-RELEASE FreeBSD 12.0-RELEASE r341666 GENERIC  amd64

root@fbsd12-ovmf:~ # dmesg | egrep -i 'scsi|mass stor|sym|sata'
sym0: <895a> port 0xb000-0xb0ff mem 0x98c02000-0x98c023ff,0x98c00000-0x98c0=
1fff
irq 22 at device 0.0 on pci2
sym0: No NVRAM, ID 7, Fast-40, LVD, parity checking
pci3: <mass storage, SCSI> at device 0.0 (no driver attached)
ahci0: <Intel ICH9 AHCI SATA controller> port 0xc240-0xc25f mem
0x99002000-0x99002fff irq 16 at device 31.2 on pci0
ada0: <QEMU HARDDISK 2.5+> ATA-7 SATA device
ada0: 150.000MB/s transfers (SATA 1.x, UDMA5, PIO 8192bytes)
cd0: <QEMU QEMU DVD-ROM 2.5+> Removable CD-ROM SCSI device
cd0: 150.000MB/s transfers (SATA 1.x, UDMA5, ATAPI 12bytes, PIO 8192bytes)

root@fbsd12-ovmf:~ # uname -a
FreeBSD fbsd12-ovmf 12.0-RELEASE-p12 FreeBSD 12.0-RELEASE-p12 r356437 GENER=
IC=20
amd64


Please note that all VirtIO connected devices still fail to load as per my =
bug
report.  Unfortunately, UEFI boot process is a bit more complicated and, th=
us,
I can't simply switch the HDD bus from SATA to SCSI :(.  When I tried a fre=
sh
install of 12.0 to the 'Hypervisor default' SCSI, the install is unable to
detect the attached SCSI HDD...  Exiting the install to shell to debug, sym
driver is already loaded in the kernel but doesn't show up dmesg.  Thus, un=
able
to troubleshoot further with installing to or booting from the SCSI HDD. :(=
  If
I boot from SATA, all non-VirtIO works as expected:

root@fbsd12-ovmf:~ # uname -a
FreeBSD fbsd12-ovmf 12.0-RELEASE-p12 FreeBSD 12.0-RELEASE-p12 r356437 GENER=
IC=20
amd64

root@fbsd12-ovmf:~ # dmesg | egrep -i 'scsi|mass stor|sym|sata|em0'
sym0: <895a> port 0xb000-0xb0ff mem 0x98c02000-0x98c023ff,0x98c00000-0x98c0=
1fff
irq 22 at device 0.0 on pci2
sym0: No NVRAM, ID 7, Fast-40, LVD, parity checking
pci3: <mass storage, SCSI> at device 0.0 (no driver attached)
em0: <Intel(R) PRO/1000 Network Connection> port 0x6000-0x603f mem
0x98000000-0x9801ffff irq 21 at device 1.0 on pci8
em0: attach_pre capping queues at 1
em0: using 1024 tx descriptors and 1024 rx descriptors
em0: allocated for 1 tx_queues
em0: allocated for 1 rx_queues
em0: Ethernet address: 52:54:00:f2:70:9c
ahci0: <Intel ICH9 AHCI SATA controller> port 0xc240-0xc25f mem
0x99002000-0x99002fff irq 16 at device 31.2 on pci0
da0 at sym0 bus 0 scbus0 target 2 lun 0
da0: <QEMU QEMU HARDDISK 2.5+> Fixed Direct Access SPC-3 SCSI device
ada0: <QEMU HARDDISK 2.5+> ATA-7 SATA device
ada0: 150.000MB/s transfers (SATA 1.x, UDMA5, PIO 8192bytes)
cd0: <QEMU QEMU DVD-ROM 2.5+> Removable CD-ROM SCSI device
cd0: 150.000MB/s transfers (SATA 1.x, UDMA5, ATAPI 12bytes, PIO 8192bytes)
em0: link state changed to UP

Perhaps something broken during the upgrade from 11.2?  Is your boot/OS disk
SATA or SCSI?  Just a FIY, not all SCSI/RAID hardware HBA supports UEFI boo=
t,
IIRC.  Thus, I don't think the emulated 'Hypervisor default' SCSI does eith=
er
:(  Here's my pciconf output when boot from SATA:

none1@pci0:1:0:0:       class=3D0x020000 card=3D0x11001af4 chip=3D0x10411af=
4 rev=3D0x01
hdr=3D0x00
    vendor     =3D 'Red Hat, Inc.'
    device     =3D 'Virtio network device'
    class      =3D network
    subclass   =3D ethernet
    bar   [14] =3D type Memory, range 32, base 0x98e00000, size 4096, enabl=
ed
    bar   [20] =3D type Prefetchable Memory, range 64, base 0x800000000, si=
ze
16384, enabled
    cap 11[dc] =3D MSI-X supports 3 messages
                 Table in map 0x14[0x0], PBA in map 0x14[0x800]
    cap 09[c8] =3D vendor (length 20)
    cap 09[b4] =3D vendor (length 20)
    cap 09[a4] =3D vendor (length 16)
    cap 09[94] =3D vendor (length 16)
    cap 09[84] =3D vendor (length 16)
    cap 01[7c] =3D powerspec 3  supports D0 D3  current D0
    cap 10[40] =3D PCI-Express 2 endpoint max data 128(128)
                 link x1(x1) speed 2.5(2.5) ASPM disabled(L0s)
sym0@pci0:2:0:0:        class=3D0x010000 card=3D0x10000000 chip=3D0x0012100=
0 rev=3D0x00
hdr=3D0x00
    vendor     =3D 'LSI Logic / Symbios Logic'
    device     =3D '53c895a'
    class      =3D mass storage
    subclass   =3D SCSI
    bar   [10] =3D type I/O Port, range 32, base 0xb000, size 256, enabled
    bar   [14] =3D type Memory, range 32, base 0x98c02000, size 1024, enabl=
ed
    bar   [18] =3D type Memory, range 32, base 0x98c00000, size 8192, enabl=
ed
none2@pci0:3:0:0:       class=3D0x010000 card=3D0x11001af4 chip=3D0x10481af=
4 rev=3D0x01
hdr=3D0x00
    vendor     =3D 'Red Hat, Inc.'
    device     =3D 'Virtio SCSI'
    class      =3D mass storage
    subclass   =3D SCSI
    bar   [14] =3D type Memory, range 32, base 0x98a00000, size 4096, enabl=
ed
    bar   [20] =3D type Prefetchable Memory, range 64, base 0x800100000, si=
ze
16384, enabled
    cap 11[dc] =3D MSI-X supports 4 messages
                 Table in map 0x14[0x0], PBA in map 0x14[0x800]
    cap 09[c8] =3D vendor (length 20)
    cap 09[b4] =3D vendor (length 20)
    cap 09[a4] =3D vendor (length 16)
    cap 09[94] =3D vendor (length 16)
    cap 09[84] =3D vendor (length 16)
    cap 01[7c] =3D powerspec 3  supports D0 D3  current D0
    cap 10[40] =3D PCI-Express 2 endpoint max data 128(128)
                 link x1(x1) speed 2.5(2.5) ASPM disabled(L0s)

BTW, my host OS is an updated Ubuntu 18.04.3 LTS.  I don't know if that has=
 any
bearing vs 19.x

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-241774-27103-UdOs41QPpz>